Skip to main content

Full text of "DTIC ADA235656: The Effects of Natural Sands on Asphalt Concrete Engineering Properties."

See other formats





Preoared for DEPARTMENT OF THE AIR FORCE 
Air Force Engineering ana Services Center 
Tyndall Air Force Base, Florida 32403-6001 














Destroy this report when no longer needed Do not return 
it to the originator 


The findings in this report are not to be construed as an official 
Department of the Army position unless so designated 
by other authorized documents 


The contents of this report are not to be used for 
advertising, publication, or promotional purposes 
Citation of trade names does not constitute an 
official endorsement or approval of the use of 
such commercial products. 





REPORT DOCUMENTATION PAGE 

Form Approved 

OMB No 0704-0188 

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, 
gathering and maintaining the data needed, and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of this 
collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for information Operations and Reports. 12IS Jefferson 
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188). Washington. DC 20503 

t. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED 

March 1991 Final report 

4 . TITLE AND SUBTITLE 

The Effects of Natural Sands on Asphalt Concrete Engineering 
Properties 

5. FUNDING NUMBERS 

6. AUTHOR(S) 

Randy C. Ahlrich 

7. PERFORMING ORGANIZATION NAME(S) AND AODRESS(ES) 

USAE Waterways Experiment Station, Geotechnical Laboratory, 

3909 Halls Ferry Road, Vicksburg, MS 39180-6199 

8. PERFORMING ORGANIZATION 

REPORT NUMBER 

Technical Report 

GL-91-3 

9. SPONSORING /MONITORING AGENCY NAME{S) AND ADDRESS(ES) 

Air Force Engineering and Services Center 

Tyndall Air Force Base, FL 32403-6001 

10. SPONSORING/MONITORING 

AGENCY REPORT NUMBER 

11. SUPPLEMENTARY NOTES 

Available from National Technical Information Service, 5285 Port Royal Road, Springfield, 

VA 22161. 

12a. DISTRIBUTION /AVAILABILITY STATEMENT 

Approved for public release; distribution unlimited 

12b. DISTRIBUTION CODE 

13. ABSTRACT (Maximum 200 words) 

Asphalt concrete rutting is premature deformation that develops in wheelpaths under channelized 
loads. Asphalt concrete mixtures that exhibit plastic flow are caused by asphalt concrete mixtures 
that have an excessive asphalt content and/or excessive amount of uncrushed rounded aggregate. 

This laboratory study was conducted to evaluate the effects of natural sands on the engineering 
properties of asphalt concrete. This research consisted of a literature review and a two-phase 
laboratory study on laboratory-produced specimens. Conventional and state-of-the-art testing 
procedures including indirect tensile, resilient modulus, and unconfined creep rebound were used to 
determine the effects of natural sands. 

The conclusions of the laboratory study indicated that use of natural sand materials decreased 
strength characteristics of asphalt concrete mixtures. Replacing natural sand materials with crushed 
aggregates increased the resistance to permanent deformation. This study recommends that the 
maximum limit for natural sand be 15 percent, but to maximize the decrease in rutting potential, all 
crushed aggregate should be used. 

14. SUBJECT TERMS 

Aggregate Indirect tensile Resilient modulus 

Asphalt concrete Natural sands Rutting 

Gvraiorv testing Permanent deformation Unconfined creeo 

15. NUMBER OF PAGES 

172 

16. PRICE CODE 

17 SECURITY ClAS'.’FICATICN 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 

OF REPORT OF THIS PAGE OF ABSTRACT 

Unclassified Unclassified Unclassified 

20. LIMITATION OF ABSTRACT 

NSN 7530 01-280-5500 Standard Form 298 (Rev 2-89) 


Pr^CftWj by ANSI StCJ *’39-'8 


798 1Q2 


















PREFACE 


This study was conducted by the Geotechnical Laboratory (GL) , 

US Army Engineer Waterways Experiment Station (WES), Vicksburg, 
Mississippi, for the Air Force Engineering and Services Center (AFESC), 
US Air Force. This work was conducted from October 1989 to December 
1990 under the project entitled, "Effects of Natural Sand on the Rutting 
of Asphalt Concrete Mixtures." The AFESC Technical Monitor was Mr Jim 
Greene. 

The study was conducted under the general supervision of Dr. W. F. 
Marcuson III, Chief, GL; Mr. H. H. Ulery, Jr., Chief, Pavement Systems 
Division (PSD), GL; and Dr. R. S. Rollings and Mr. L. N. Godwin, former 
Chiefs, Materials Research and Construction Technology (MRCT) Branch, 
PSD. This report was produced under direct supervision of Mr. T. W. 
Vollor, Acting Chief, MRCT Branch, PSD. Personnel engaged in the 
testing, evaluating, and analysis of this project included Messrs. 

T. Carr, J. Duncan, R. Graham, T. McCaffrey, H. McKnight, J. Simmons, 
and D. Reed. The Principal Investigator was Mr. R. C. Ahlrich who also 
wrote this report. 

COL Larry B. Fulton, EN, is Commander and Director of WES. 

Dr. Robert W. Whalin is Technical Director. 







TABLE OF CONTENTS 


PAGE 

PREFACE . i 

LIST OF TABLES. iv 

LIST OF FIGURES. v 

CONVERSION FACTORS, NON-SI TO SI (METRIC) UNITS OF MEASUREMENT. . . ix 

CHAPTER 

I. INTRODUCTION. 1 

Background. 1 

Purpose . 3 

Objective . 3 

Scope. 4 

II. REVIEW OF LITERATURE. 8 

III. DISCUSSION AND DESCRIPTION OF TESTING 

EQUIPMENT AND PROCEDURES. 14 

Gyratory Testing Machine. 14 

Automated Data Acquisition Testing System . 20 

Indirect Tensile. 22 

Resilient Modulus . 25 

Unconfined Creep-Rebound. 28 

IV. PHASE I - PRESENTATION AND ANALYSIS OF DATA. 33 

Asphalt Cement. 33 

Natural Sand Materials. 34 

Limestone Aggregate . 36 

Aggregate Blends. 36 

Mix Designs. 46 

Mixture Properties at Optimum Asphalt Content . 61 

V. PHASE II - PRESENTATION AND ANALYSIS Or D«iA. 6' 





























V. PHASE II - PRESENTATION AND ANALYSIS OF DATA. 66 

Marshall Mix Properties .. 67 

Indirect Tensile. 77 

Resilient Modulus . 7 g 

Unconfined Creep-Rebound. 97 

chapter page 

VI. SUMMARY. CONCLUSIONS, AND RECOMMENDATIONS . 110 

Summary.HO 

Conclusions.. 

Recommendations.. 

REFERENCES.. 

APPENDIX A: UNCONFINED CREEP-REBOUND CURVES.120 


iii 














LIST OF TABLES 


TABLE PAGE 

1 Natural Sand Laboratory Study Test Plan . 7 

2 Asphalt Cement Properties . 34 

3 Aggregate Gradations for Natural Sands. 35 

4 Aggregate Gradations for Natural Sand Laboratory Study. ... 38 

5 Criteria for Determining Acceptability of Mixture . 47 

6 Asphalt Concrete Mix Design Properties. 50 

7 Summary of Mix Design Properties at Optimum Asphalt Content . 62 

8 Summary of Mix Properties at Optimum Asphalt Content. 68 

9 Marshall Stability and Flow Results . 72 

10 Summary of Marshall Stability Values. 75 

11 Indirect Tensile Test Results . 79 

12 Summary of Indirect Tensile Test at 77°F .83 

13 Summary of Indirect Tensile Test at 104°F .84 

14 Resilient Modulus Test Results. 88 

15 Summary of Resilient Modulus Test Results . 95 

16 Unconfined Creep-Rebound Test Results . 101 

17 Summary of Unconfined Creep-Rebound Test at 77°F 105 

18 Summary of Unconfined Creep-Rebound Test at 104°F 106 

19 Summary of Creep Modulus Values . 108 

iv 






















LIST OF FIGURES 


FIGURE PAGE 

1 Flow Diagram of Natural Sand Laboratory Study . 6 

2 Schematic of Gyratory Compaction Process . 16 

3 WES Model 4C Gyratory Testing Machine . 17 

4 Typical Gyrograph . 19 

5 Overall View of Automated Data Acquisition Testing System . . 21 

6 Schematic of Indirect Tensile Test. 23 

7 Indirect Tensile Test . 24 

8 Resilient Modulus Test. 26 

9 Unconfined Creep-Rebound Test . 30 

10 Typical Unconfined Creep-Rebound Curve. 31 

11 Aggregate Gradation Curve for S-1M Blend. 39 

12 Aggregate Gradation Curve for S-2M Blend. 39 

13 Aggregate Gradation Curve for S-3M Blend. 40 

14 Aggregate Gradation Curve for S-1C Blend. 40 

15 Aggregate Gradation Curve for S-2C Blend. 41 

16 Aggregate Gradation Curve for S-3C Blend. •. 41 

17 S-0 Gradation Curve Raised to 0.45 Power. 42 

18 S-1M Gradation Curve Raised to 0.45 Power. 42 

19 S-2M Gradation Curve Raised to 0.45 Power. 43 

20 S-3M Gradation Curve Raised to 0.45 Power. 43 


v 






















FIGURE PAGE 

21 S-1C Gradation Curve Raised to 0.45 Power. 44 

22 S-2C Gradation Curve Raised to 0.45 Power. 44 

23 S-3C Gradation Curve Raised to 0.45 Power. 45 

24 Marshall Apparatus . 48 

25 Mix Design Plots for S-0 Blend. 54 

26 Mix Design Plots for S-1M Blend. 55 

27 Mix Design Plots for S-2M Blend. 56 

28 Mix Design Plots for S-3M Blend. 57 

29 Mix Design Plots for S-1C Blend. 58 

30 Mix Design Plots for S-2C Blend. 59 

31 Mix Design Plots for S-3C Blend. 60 

32 Optimum Asphalt Content Versus Percent Mason Sand . 63 

33 Optimum Asphalt Content Versus Percent Concrete Sand. 63 

34 Marshall Stability Versus Percent Mason Sand. 64 

35 Marshall Stability Versus Percent Concrete Sand . 64 

36 Voids in Mineral Aggregate Versus Percent Mason Sand. 65 

37 Voids in Mineral Aggregate Versus Percent Concrete Sand ... 65 

38 VMA Versus Percent Mason Sand. 70 

39 VMA Versus Percent Concrete Sand. 70 

40 Marshall Stability Versus Percent Mason Sand. 76 

41 Marshall Stability Versus Percent Concrete Sand . 76 

42 Indirect Tensile Strength Values at 77°F 85 

43 Indirect Tensile Strength Values at 104°F . 85 

44 Resilient Modulus Values at 77°F 96 

vi 





























FIGURE PAGE 

45 Resilient Modulus Values at 104°F . 96 

46 Axial Deformation Values at 77°F.107 

47 Permanent Deformation Values at 77°F 107 

48 Creep Modulus Values at 77°F 109 

49 Creep Modulus Values at 104°F 109 

50 Creep-Rebound Curve for S-01 Sample at 77°F 121 

51 Creep-Rebound Curve for S-02 Sample at 77°F 122 

52 Creep-Rebound Curve for S-03 Sample at 77°F 123 

53 Creep-Rebound Curve for S-04 Sample at 104°F.124 

54 Creep-Rebound Curve for S-05 Sample at 104°F.125 

55 Creep-Rebound Curve for S-06 Sample at 104°F.126 

56 Creep-Rebound Curve for S-1M1 Sample at 77°F 127 

57 Creep-Rebound Curve for S-1M2 Sample at 77°F.128 

58 Creep-Rebound Curve for S-1M3 Sample at 104°F 129 

59 Creep-Rebound Curve for S-1M4 Sample at 104°F 130 

60 Creep-Rebound Curve for S-1M5 Sample at 104°F 131 

61 Creep-Rebound Curve for S-2M1 Sample at 77°F.132 

62 Creep-Rebound Curve for S-2M2 Sample at 77°F.133 

63 Creep-Rebound Curve for S-2M3 Sample at 104°F 134 

64 Creep-Rebound Curve for S-2M4 Sample at 104°F 135 

65 Creep-Rebound Curve for S-2M5 Sample at 104°F 136 

66 Creep-Rebound Curve for S-2M6 Sample at 104°F 137 

67 Creep-Rebound Curve for S-3M1 Sample at 77°F 138 

68 Creep-Rebound Curve for S-3M2 Sample at 77°F.139 

vii 





























FIGURE PAGE 

69 Creep-Rebound Curve for S-3M3 Sample at 77°F 140 

70 Creep-Rebound Curve for S-3M4 Sample at 77°F.141 

71 Creep-Rebound Curve for S-3M5 Sample at 104°F 142 

72 Creep-Rebound Curve for S-3M6 Sample at 104°F 143 

73 Creep-Rebound Curve for S-3M7 Sample at 104°F 144 

74 Creep-Rebound Curve for S-1C1 Sample at 77°F. 145 

75 Creep - Rebound curve for S-1C2 Sample at 77°F. 146 

76 Creep-Rebound curve for S-1C3 Sample at 104°F 147 

77 Creep-Rebound Curve for S-1C4 Sample at 104°F 148 

78 Cree^-Rebound Curve for S-1C5 Sample at 104°F 149 

79 Creep-Rebound Curve for S-2C1 Sample at 77°F. 150 

80 Creep-Rebound Curve for S-2C2 Sample at 77°F. 151 

81 Creep - Rebound Curve for S-2C3 Sample at 77°F. 152 

82 Creep Rebound Curve for S-2C4 Sample at 104°F 153 

83 Creep - Rebound Curve for S-2C5 Sample at 104°F 154 

84 Creep - Rebound Curve for S-2C6 Sample at 104°F 155 

85 Creep-Rebound Curve for S-3C1 Sample at 77°F. 156 

86 Creep-Rebound Curve for S-3C2 Sample at 77°F. 157 

8 7 Creep - Rebound Curve for S-3C3 Sample at 77°F. 158 

88 Creep - Rebound Curve for S-3C4 Sample at 104°F 159 

89 Creep-Rebound Curve for S-3C5 Sample at 104°F 160 

90 Creep - Rebound Curve for S-3C6 Sample at 104°F 161 


vi i i 





























CONVERSION FACTORS, NON-SI TO SI (METRIC) 
UNITS OF MEASUREMENT 


Non-SI units of measurement used in this report can be converted to SI 
(metric) units as follows: 


Multiply 

Bv 

To Obtain 

degrees (angle) 

0.01745329 

radians 

Fahrenheit degrees 

5/9 

Celsius degrees or Kelvins * 

inches 

2.54 

centimetres 

pounds (force) 

4.448222 

newtons 

pounds (force) per 
square inch 

6.894757 

kilopascals 

pounds (mass) per 
cubic foot 

16.01846 

kilograms per cubic metre 


* To obtain Celsius (C) temperature readings from Fahrenheit (F) 

readings, use the following formula: C - (5/9)(F - 32). To obtain 
Kelvin (K) readings, use: K - (5/9)(F - 32) + 273.15. 


ix 




The Effects of Natural Sands on Asohal t- 


Concrete Engineering Properties 


CHAPTER I 
INTRODUCTION 
Background 

In recent years, deterioration of asphalt concrete pavements on 
military installations and state highways has increased. This 
deterioration has been caused by higher traffic volumes, higher traffic 
loads, increasing tire pressures, poor construction quality control and 
decreased quality of asphalt concrete mixtures. Rutting is one of the 
most common forms of deterioration in asphalt concrete pavement (6). 

Asphalt concrete rutting is generally premature longitudinal 
deformation that develops in the wheelpaths under channelized loads. 
Rutting of asphalt concrete pavements is a complicated process and can 
be caused by several factors. Rutting is typically caused by one of the 
following: 1) shear deformation of base course or subgrade, 2) densifi- 

cation or consolidation of base course or subgrade, 3) densification or 
consolidation of asphalt concrete material, and U) plastic flow of 
asphalt concrete material (2). 

Rutting of an asphalt concrete pavement caused by plastic flow of 
the asphalt concrete material indicates a problem with the asphalt 
concrete mixture. Plastic flow of an asphalt concrete material illus¬ 
trates an unstable mixture. Rutting of this nature is demonstrated by a 
depression under the loaded area with humps on either side. Asphalt 


1 








2 


concrete mixtures that exhibit plastic flow rutting are generally caused 
by asphalt concrete mixtures that have an excessive asphalt content, an 
improper gradation, and/or an excessive amount of uncrushed rounded 
aggregate. 

Uncrushed rounded aggregates have been proven to decrease the 
strength properties of asphalt concrete mixtures and produce materials 
that are unstable. Natural sand materials, which are primarily 
uncrushed rounded particles, are often used in asphalt concrete mixtures 
because these materials are generally less expensive, readily available, 
and can be blended easily with other materials. Natural sand materials 
have a smooth, rounded surface texture that greatly reduces the inter¬ 
locking properties of the asphalt concrete and reduces the strength 
properties. Low strength properties and stability values in asphalt 
concrete mixtures allow deformation to occur, which leads to rutting 
(18,20,22,24). 

In numerous field evaluations by the Federal Highway Administration 
(FHWA) and U.S. Corps of Engineers of asphalt concrete pavements that 
had exhibited rutting, it was found that many highway departments and 
military installations were allowing an excess of natural sand in their 
asphalt concrete mixtures. This excessive amount of uncrushed rounded 
particles was causing a reduction in pavement strength and stability and 
an increase in permanent deformation under traffic (2,5,19,27,31). 

Most agencies that construct flexible pavements have some guidance 
or have set allowable limits on the use ol natural sands. The Corps of 
Engineers has set allowable limits for natural sand content for asphalt 
concrete mixtures, but these limits are not widely used outside major 







3 

airfield paving projects (9). Some state highway departments have also 
set limits for the amount of natural sand, but the maximum limit varies 
from 10 to 30 percent. Some other highway departments have no limits 
and allow an unlimited amount of natural sand. The general consensus is 
that the maximum limit for natural sand is not generally controlled. 

The natural sand limits established by the Corps of Engineers are 
based on past observed behavior and performance in the field. Labora¬ 
tory evaluations have not been conducted to determine allowable limits 
for natural sands. Since the widely specified Marshall mix design 
procedure does not always reflect the detrimental effect of natural 
sand, many mix designs produced for state highway departments and 
military installations have an excess amount of natural sand. 

Purpose 

The purpose of this research was to evaluate the effects of natural 
sands on the engineering properties of asphalt concrete. This research 
provided a sound basis for selecting allowable natural sand contents for 
asphalt concrete mixtures to increase strength and stability and 
decrease the rutting potential. The documentation of this work provided 
strong support for the wide use of natural sand content limics in 
asphalt concrete mixtures. This information has the potential to 
improve the rutting performance of asphalt concrete at a negligible 
additional cost compared to other more costly approaches such as asphalt 
binder modifiers. 

Obiective 

The objective of this research is to determine the influence of 
various amounts of natural sands on the engineering properties of 




4 


asphalt concrete mixtures and to set quantitative limits of natural sand 
to prevent unstable mixtures and reduce rutting potential. 

Scope 

The scope of this research study included a review of available 
literature and existing data, a two-phase laboratory study on 
laboratory-produced samples, and an analysis of the data. Both 
conventional and state-of-the-art testing procedures were incorporated 
into the laboratory test plan. Asphalt concrete mixture tests that were 
performed included the Marshall stability and flow, indirect tensile, 
resilient modulus and unconfined creep - rebound tests. A diagram of the 
laboratory test plan used in this study is shown in Figure 1. 

To evaluate the effect of natural sands on asphalt concrete 
mixtures, a laboratory study was conducted using two gradations of 
natural sand material with four different percentages of sand in the 
asphalt concrete mixtures. The asphalt concrete mixtures were produced 
with 0, 10, 20, and 30 percent natural sand. Each aggregate blend was 
fabricated in the laboratory with a constant mixture gradation. 

The test plan for the natural sand laboratory evaluation is 
summarized in Table 1. Phase I of the laboratory study involved testing 
the laboratory materials (aggregates, sands, and asphalt cement) and 
conducting mix designs for the seven asphalt concrete mixtures using the 
Marshall mix design criteria. All asphalt concrete samples were 
compacted with the Corps of Engineers Gyratory Testing Machine (GTM) 
using 200 psi pressure, 1-degree gyration angle, and 30 revolutions 
which is equivalent to the 75-blow Marshall hand hammer compactive 







5 

effort. The optimum asphalt content for each aggregate blend was 
selected at 4 percent voids total mix in the asphalt concrete mixtures. 

Phase II of the laboratory study involved conducting a series of 
laboratory tests to determine the engineering properties of the seven 
asphalt concrete mixtures. Forty specimens at the optimum asphalt 
content were produced for each aggregate blend. The following 
laboratory tests were conducted on the specimens: 

1. Marshall stability and flow at 140°F. 

2. Indirect tensile at 77°F and 104°F. 

3. Resilient modulus at 77°F and 104°F. 

4. Unconfined creep - rebound at 77°F and 104°F. 

Several repetitions of each test were performed in order to provide 
sufficient data for a complete analysis. A total of 280 specimens were 
analyzed. From this series of tests, the effects of natural sands on 
the engineering properties were determined. 






Figure 1 . 


Flow Diagram of Natural Sand Laboratory Study 























7 


TABLE 1 

NATURAL SAND LABORATORY STUDY TEST PLAN 


1. Phase I - Material Evaluation and Mix Designs 

a. Test laboratory materials - aggregates, sands, asphalt cement 

b. Select natural sand materials - mason and concrete 

c. Produce aggregate blends for various percentages of sand - 
0, 10, 20, 30 

d. Conduct seven asphalt concrete mix designs with laboratory 
limestone labstock and two natural sands 

e. Compact all asphalt concrete specimens with Gyratory Testing 
Machine (GTM) 

f. Select optimum asphalt content at 4 percent voids total mix 


2. Phase II - Laboratory Evaluation 

a. Produce 40 specimens at optimum asphalt content for seven 
aggregate blends - Total of 280 specimens 

b. Designations for seven aggregate blends 


Blend 

Material 

S-0 

100 percent limestone 


S-1M 

90 

percent 

limestone 

- 10 

percent 

mason 

sand 

S-2M 

80 

percent 

limestone 

- 20 

percent 

mason 

sand 

S-3M 

70 

percent 

limestone 

- 30 

percent 

mason 

sand 

S-1C 

90 

percent 

limestone 

- 10 

percent 

concrete 

sand 

S-2C 

80 

percent 

1 imestone 

- 20 

percent 

concrete 

sand 

S-3C 

70 

percent 

limestone 

- 30 

percent 

concrete 

sand 


c. Conduct the following test on each aggregate blend 

1. Marshall stability and flow at 140°F 

2. Indirect tensile at 77°F and 104°F 

3. Resilient modulus at 77°F and 104°F 

4. Unconfined creep-rebound at 77°F and 104°F 





CHAPTER II 


REVIEW OF LITERATURE 


One of the most serious problems affecting our road system today is 
rutting of asphalt concrete pavements. For the last 15 years, state 
highway departments throughout the country have reported an increase in 
premature rutting (2). Many studies and evaluations have been conducted 
to determine the causes of rutting. During the development of the 
Marshall procedure at the Waterways Experiment Station (WES)(21), 
evaluations indicated that the characteristics of the fine aggregate 
control the capacity of dense-graded asphalt concrete mixtures to resist 
traffic-induced stresses that cause rutting. 

Brown (6) indicated several factors contributed to the potential 
problems that produce rutting. The factors listed included excessive 
asphalt content, excessive use of natural sand, improperly crushed 
aggregate, and low field density. Laboratory studies and field 
evaluations conducted in the states of Wyoming (31), New Mexico (19), 
and Florida (27) also identified excessive sand-size particles and 
rounded aggregates as two factors that caused rutting in asphalt 
concrete pavements. 

Numerous laboratory research studies have been conducted comparing 
crushed coarse and fine aggregates to natural or uncrushed aggregates in 
asphalt concrete mixtures. Many of the laboratory evaluations were 


8 











9 


performed during the 1950's and 1960's. Herrin and Goetz (20) evaluated 
the effect of aggregate shape on the stability of asphalt concrete 
materials. This research involved crushed and uncrushed gravel, crushed 
limestone for the coarse aggregate, and natural sand and crushed lime¬ 
stone sand for the fine aggregate. The primary conclusion was that the 
strength of the mixture, regardless of the type of coarse aggregate, 
increased substantially when fine aggregate was changed from rounded 
natural sand to crushed limestone. A secondary conclusion was that the 
strength of the mixture was affected more by a change in the fine 
aggregate than a change in the coarse aggregate. 

In 1961, Wedding and Gaynor (30) researched the effect of aggregate 
particle shape in well-graded asphalt concrete mixtures. The percent¬ 
ages of crushed coarse aggregates and the types of fine aggregates which 
included natural and washed concrete sands were varied in the mixtures. 
Comparisons of these different aggregate blends were conducted on 
specimens produced using the Marshall procedure. Mixtures with crushed 
aggregates produced higher stability values than mixtures with 
uncrushed, rounded aggregates. The substitution of all crushed 
aggregate for natural sand and gravel also increased the stability 
approximately 45 percent. 

Griffith and Kallas (17,18) researched the effects of aggregate 
types on void and strength characteristics of asphalt concrete mixtures. 
Uncrushed gravel mixtures were found to develop voids lower than the 
voids in crushed aggregates mixtures. Griffith and Kallas also 
evaluated the influence of fine aggregates on the strength of asphalt 
concrete specimens. Combinations of aggregate blends with natural and 



10 


crushed coarse aggregate and natural sand fine aggregate were analyzed. 
An increase in angularity or crushed faces increased the Hveem and 
Marshall stability values at optimum asphalt content. An increase in 
angularity in the fine aggregates also increased the minimum void 
percentages and increased optimum asphalt contents. 

Shklarsky and Livneh (29) conducted a study evaluating the 
difference between uncrushed and crushed coarse aggregate combined with 
natural sand and crushed fine aggregate. Replacing natural sand 
materials with crushed fine aggregate increased the stability and 
strength properties in Marshall specimens and reduced permanent 
deformation, improved resistance to water, reduced asphalt cement 
sensitivity, and increased voids. Shklarsky and Livneh also concluded 
that replacing uncrushed coarse aggregate with crushed material did not 
significantly improve the asphalt concrete mixture. 

Kalcheff and Tunnicliff (22) researched the effects of coarse 
aggregate gradations, shape effects of fine aggregates, and effects of 
high mineral filler content. Asphalt concrete specimens were produced 
using the Marshall and Hveem procedures with aggregate blends composed 
of natural and manufactured (crushed) sands. The optimum asphalt 
content was approximately the same for natural sand mixtures and 
manufactured sand mixtures if the sands had similar particle shape. The 
optimum asphalt content would be higher if the manufactured sand had 
more angular particles. Also, mixtures containing crushed coarse and 
fine aggregates were more resistant to permanent deformation from 
repeated traffic loadings, and much less susceptible to the effects of 
temperature than comparable mixtures containing natural sand. 











11 


Button and Perdomo (8) conducted a laboratory study that was 
designed to evaluate the effects of natural sands on permanent 
deformation and o quantify the influence on resistance to plastic 
deformation when natural sand is replaced with crushed aggregate. 
Increases in total deformation occurred as the percentage of natural 
sand increased. The texture, shape, and porosity of the fine aggregate 
were major factors controlling plastic deformation in asphalt concrete 
mixtures. They recommended replacing the natural sand material with 
manufactured sand to increase the resistance of the asphalt concrete 
pavement to permanent deformation. 

Marks, Monroe, and Adam (24) conducted a laboratory evaluation that 
analyzed the effects of crushed particles in asphalt concrete mixtures. 
Mixtures at various percentages of crushed material were evaluated. 
Laboratory testing included the Marshall stability, indirect tensile, 
resilient modulus, and creep tests. Increased percentages of crushed 
material yielded a substantial increase in stability. Resilient modulus 
data did not correlate with the percent of crushed particles or indicate 
resistance to rutting. Data from the creep test indicated rutting 
potential was very dependent on the percent of crushed aggregate. 

Marker (23) stated that natural sands or uncrushed aggregate passing 
the No. 4 sieve was the most important factor contributing to tenderness 
of an asphalt concrete mixture. Most tender pavements have an excess of 
middle-sized sand particles in the aggregate gradation. A hump in the 
grading curve that has the sieve sizes raised to the 0.45 power is 
caused by the excess sand and occurs between the No. 4 and No. 100 
sieves (11). Tenderness is most critical when this hump is near the 



12 


No. 30 sieve. This condition is generally accompanied by a relatively 
low amount of material passing the No. 200 sieve. Marker also stated 
that rounded, uncrushed aggregates are more likely to contribute to 
tender mixes than angular, crushed particles. This is especially true 
for the material passing the No. 4 sieve. 

Grau (16) demonstrated in field test sections that increases in 
amounts of natural sand and finer sand gradations produced less stable 
asphalt concrete mixtures. The asphalt mixtures progressively weakened 
under traffic as the pavement temperatures increased. A large decrease 
in stability occurred when natural gravel and sand were used together. 
The stability values of the asphalt mixtures increased significantly 
when a crushed sand was used in place of natural sand. 

The AASHTO Joint Task Force on rutting (2) reported that some 
deficiencies that have been identified as causes of rutting in asphalt 
concrete pavements include improper aggregate gradation and excessive 
use of rounded aggregates. The Task Force recommended that clean, hard 
and angular aggregates be used in asphalt concrete mixtures for high 
volume roads to help resist rutting. The FHWA Technical Advisory 
5040.27 (14) recommended that natural sands be limited to 15 to 20 per¬ 
cent of the total weight of the aggregate for high volume roads. It was 
also recommended that agencies experiencing rutting problems should 
consider reducing the use of natural sands and incorporating more 
crushed fines into their mixtures. 

In 1984, the Western Association of State Highway and Transportation 
Officials (WASHT0)(32) stated that "rutting is the most pressing issue 
facing highway agencies". WASHTO also stated "that state Materials 









13 


Engineers do not feel that the present procedures and specifications 
fully address the rutting problem. The general feeling is that the 
present state-of-the-art in materials testing relating to rutting needs 
to be upgraded through basic research". 






CHAPTER III 


DISCUSSION AND DESCRIPTION OF TESTING EQUIPMENT AND PROCEDURES 

Several types of testing equipment and test procedures were used to 
determine the effects of natural sands on the engineering properties of 
asphalt concrete. Current state-of-the-art testing equipment was used 
in addition to standard laboratory equipment and procedures generally 
used to conduct Marshall mix designs. This more complex testing 
equipment and sophisticated testing procedures included the Corps of 
Engineers Gyratory Testing Machine (GTM), Automated Data Acquisition 
Testing (ADAT) System, indirect tensile cest, resilient modulus test, 
and unconfined creep - rebound test. The laboratory equipment and test 
procedures used in this study are described and discussed in the 
following paragraphs. 

Gyratory Testing Machine 

Compaction of asphalt concrete materials using gyratory method 
applies normal forces to both the top and bottom faces of the material 
confined in cylindrically-shaped molds. Normal forces at designated 
pressures are supplemented with a kneading action or gyratory motion to 
compact the asphalt concrete material into a denser configuration while 
totally confined. The U.S. Army Corps of Engineers has developed a 
method, procedure, and equipment using this compaction procedure (11, 
25 , 26 ) . 


14 








15 


The gyratory compaction method involves placing asphalt concrete 
material into a 4-inch-diameter mold and loading into the GTM at a 
prescribed normal stress level which represents anticipated traffic 
contact pressure. The asphalt material and mold are then rotated 
through a 1-degree gyration angle for a specified number of revolutions 
of the roller assembly. Figure 2 is a schematic of the gyratory 
compaction process. Military Standard 620 A Method 102 has correlated 
equivalent types of compaction and compactive efforts (12). 

Marshall 

Gyratory Compaction Impact Compaction 

100 psi, 1-degree, 30 revolutions 50 blow per side 

200 psi, 1-degree, 30 revolutions 75 blow per side 

A Model 4C Gyratory Testing Machine (GTM) was used to compact all 
laboratory specimens in the natural sand laboratory study. Previous 
research with the GTM has suggested that the laboratory tests will 
simulate field behavior and performance under traffic when asphalt 
mixtures are compacted at stress levels similar to anticipated field 
traffic conditions (21,28). The gyratory compactive effort used in this 
laboratory evaluation followed the standard guidance in Military 
Standard 620A for the 75-blow compactive effort. The gyratory 
compactive effort was set at the 200 psi normal stress level, 1-degree 
gyration angle, and 30 revolutions of the roller assembly. The asphalt 
concrete specimens produced with this compactive effort satisfied the 
Marshall specimen dimensions of 4 inches in diameter and 2 1/2 inches 
thick. Figure 3 shows the WES Model 4C GTM. 






6 



!' 1 f’.ure ? . Schemat i r oi Gyratory Compaction Process 

















18 

The gyratory compaction method using the GTM produces a gyratory 
graph or gyrograph that can be used to evaluate the asphalt concrete 
mixture behavior during compaction. The gyrograph indicates the 
relative stability behavior of the mixture during the compactive effort. 
The gyrograph indicates an unstable mixture when the gyrograph spreads 
or widens. A gyrograph that does not spread is considered stable under 
that loading condition (25,26). 

The gyrograph can be used to produce two indices that describe the 
relative stability of an asphalt concrete mixture. The ratio of the 
final width to the intermediate width of the gyrograph is called the 
Gyratory Stability Index (GS1). A GSI value greater than 1.0 indicates 
an unstable mixture with a high asphalt content. The ratio of the 
intermediate width to the initial width is called the Gyratory Elasto- 
Plastic Index (GEPI). The GEPI value is an indicator of the quality of 
the aggregate. Figure 4 displays a typical gyrograph of a compacted 
asphalt concrete specimen. 









19 


CHART NO. 



0 O = Initial gyratory angle (divisions) 
Q i = Intermediate gyrograph width 
0 max = Maximum gyrograph width 


Figure 4. Typical Gyrograph (aftci McRae, 1965) 



















20 


Automated Data Acquisition Testing System 

Previous research studies conducted in the Materials Research and 
Construction Technology Branch, Geotechnical Laboratory, had required 
accurately controlled laboratory testing and data acquisition (4). A 
state-of-the-art computer-operated system was assembled to conduct 
modern, complex asphalt concrete mixture tests. This customed-designed 
computer-testing system is called the Automated Data Acquisition Testing 
(ADAT) System. The ADAT System was specifically designed and organized 
to conduct three asphalt concrete mixture tests; indirect tensile, 
resilient modulus, and unconfined creep-rebound. Figure 5 is an overall 
view of the ADAT System. 

The MTS electrohydraulic closed-looped material system is the main 
component of the ADAT System. The loading sequences of the electro - 
hydraulic system are controlled by an arbitrary waveform generator. The 
test loads are recorded by electronic load cells and the specimen 
deformations are measured by electronic linear variable differential 
transformers (LVDT). The ADAT System also includes electronic 
temperature control of the enclosed environmental chamber and real time 
color graphics. 

The ADAT System is controlled by a 16-bit mini-computer designed to 
operate as the system's principal measurement and control station. 
Customized computer programs were developed to control the mechanics, 
monitoring systems, test daf°. manipulations, and data storage for 
indirect tensile, resilient modulus and unconfined creep-rebound tests. 
These programs were designed to reduce operator dependency and to allow 
the computer to be the single system control. 
















22 


Indirect Tensile 

Researchers in Brazil and Japan developed a testing procedure in 
1953 to indirectly determine tensile strengths of materials (1). The 
indirect tensile test involves placing a cylinder of material 
horizontally between two loading plates and loading the specimen across 
its diameter until failure. This test procedure has been used to test 
soils, concrete, and asphalt concrete materials, and has been used by 
engineers to compute fundamental properties of materials. Figure 6 
shows a schematic of the indirect tensile test. 

ASTM Method D4123 provides guidance on indirect tensile testing of 
asphalt concrete mixtures (3). This test procedure was conducted on 
specimens produced at the optimum asphalt content for each aggregate 
blend. This test procedure is considered straight forward and generally 
produces consistent results. The indirect tensile test was conducted on 
specimens at two test temperatures, 77°F and 104°F. These specimens 
were cured in an oven at the appropriate temperature for 24 hours before 
testing in the environmental chamber of the ADAT System. 

The indirect tensile test required that the specimens be positioned 
so that the loading plates were centered and the load was applied across 
the diameter of the specimen. The vertical load was applied at a 
constant deformation rate of 2 inches per minute until failure. The 
ultimate load was recorded at failure by the ADAT System and used to 
calculate the tensile strength. This testing procedure was conducted on 
a minimum of three specimens for each of the seven aggregate blends at 
both temperatures. Figure 7 shows the indirect tensile test. 




























25 


The tensile strength was calculated using the formulation provided 
in ASTM D4123, as follows: 

Tensile strength = 2P/?rtD 

where 

P = ultimate load required to fail specimen (lb) 
t = thickness of specimen (in) 

D = diameter of specimen (in) 

The results of the indirect tensile tests are presented and discussed in 
Chapter V. 

Resilient Modulus 

The resilient modulus test is used to evaluate the relative quality 
of asphalt concrete mixtures. The resilient modulus test procedure was 
conducted according to ASTM Method D4123 (3). Higher resilient modulus 
values indicate that the asphalt mixture has a greater resistance to 
permanent elastic deformation. This test procedure also evaluates the 
effects of repeated loads on asphalt concrete mixtures. The resilient 
modulus test is considered a nondestructive test and allows the same 
specimen to be tested several times. 

The resilient modulus test requires the specimens to be pre¬ 
conditioned at the desired testing temperature for 24 hours. The 
specimens are then positioned between the loading plates in the same 
manner as the indirect tensile test. Horizontal and vertical 
deformations are measured during the loading operation with LVDTs. 


Figure 8 shows the resilient modulus test. 







26 



Figure 8. Resilient Modulus Test 












27 


The actual resilient modulus testing procedure for this study 
involved the following: the specimens were preconditioned by applying a 
repeated haversine waveform at a reduced load to obtain a uniform 
deformation readout; the magnitude of the load applied was 5 to 25 per¬ 
cent of the aggregate blend's tensile strength; the time of loading was 
set at 0.1 seconds (representative time for actual pavement loadings); 
the loading frequency was set at 1.0 Hz or 1 cycle per second; and the 
haversine waveform was applied by the arbitrary waveform generator as 
recommended by ASTM. 

The resilient modulus test was conducted on a minimum of six 
specimens from each aggregate blend. Each specimen was tested in two 
positions, the initial position (0 degrees) and a rotated position 
90 degrees from the initial position. Conducting the resilient modulus 
test in this manner allowed a total of twelve resilient modulus values 
to be determined. This procedure was conducted at both testing 
temperatures, 77°F and 104°F. 

The resilient modulus value was calculated using a modified version 
of the equation presented in ASTM D4123. The equation used in this 
study assumed a Poisson's ratio of 0.35. The ASTM method suggests an 
equation that uses a Poisson's ratio that is calculated with horizontal 
and vertical deformations. The variability in the measured vertical 
deformation causes an inconsistency in the calculated resilient modulus 
value, thus producing unreliable data (7). 





28 


The resilient modulus value was calculated as follows: 

E rt = 0.62P/t AH t (4) 

where 

E rt -= total resilient modulus of elasticity (psi) 

P = applied repeated load (lb) 
l = thickness of’ specimen (in) 

AH t total recoverable horizontal deformation (in) 

The results of the resilient modules cests are presented and discussed 
in Chapter V. 

Uncc ' ifined Creep-Rebound 

The unconfined creep-rebound test used to evaluate the natural sand 
' aggregate blends was developed at WES (4). This test method has no 

nationally recognized test procedure. The unconfined creep-rebound test 
was developed to evaluate the asphalt mixture's resistance to permanent 
deformation under severe loads. This laboratory test is one of the best 
indicators of rutting potential. The rebound portion of the test 
procedure evaluates the reaction of the asphalt concrete after severe 
loading. 

The unconfined creep - rebound tests were performed on three Karshal1 
specimens stacked on top of each other. These specimens were 
approximately 7 1/2 inches tall. ' r he specimens were placed in the 
environmental chamber between the loading plates after curing in the 
oven for 24 hours. The loading plates were precoated with silicone 
grease to minimize the effect of end restraint. Two vertical LVDTs were 
mounted on the center specimen to record the vertical deformation during 
the loading and unloading phases. 


An average of the two readings were 











29 


used to make the creep-rebound calculations. Each stack of specimens 
was preconditioned with a 50-pound preload, approximately a 4 psi 
vertical stress, before the actual testing began. Figure 9 shows the 
unconfined creep - rebound test. 

The creep portion of the test applied a constant load for 60 minutes 
and then the load was released for 60 minutes for the rebound phase. 

The deformations and loads were recorded by the ADAT System at various 
times during the creep and rebound phases. These measurements were used 
to calculate stresses and strains and then converted into a creep 
modulus value. The unconfined creep-rebound test was conducted at 77°F 
and 104°F. The constant loads applied to the specimens ranged from 30 
to 40 psi for the 77°F tests and 10 to 15 psi for the 104°F tests. 

Figure 10 displays a typical creep-rebound deflection versus time curve. 

The results of the unconfined creep-rebound test can be used in 
several ways to evaluate asphalt concrete mixtures. The amount of 
deformation during the creep phase indicates the asphalt mixture's 
potential resistance to permanent deformation. Smaller axial 
deformations and lower creep deformation values indicate stable asphalt 
mixtures. The percent rebound or recovered deformation indicates the 
asphalt concrete mixture's ability to recover traffic induced deforma¬ 
tion. High percent rebound values indicate that little deformation will 
actually occur. The creep modulus value indicates the asphalt concrete 
mixture's stiffness. High creep modulus values should indicate minimum 
potential permanent deformation. 













UNCONFINED CREEP-REBOUND 



Figure 10. Typical Unconfined Creep-Rebound Curve 









32 


The creep modulus value was calculated as follows: 

Ec - (S)(H)/D (4) 

where 

Ec - creep modulus (psi) 

S - vertical stress (load/contact area; psi) 

H - height of specimen (in) 

D - axial deformation (in) 

Test results for the creep, rebound and creep modulus values are 
presented and discussed in Chapter V. 








CHAPTER IV 

PHASE I - PRESENTATION AND ANALYSIS OF DATA 

This chapter presents and discusses the results of the laboratory 
testing involved in Phase I of this laboratory study. Laboratory tests 
were conducted on the laboratory materials to determine physical 
properties of the asphalt cement, natural sand materials, and labstock 
limestone aggregate. Aggregate gradations were computed to produce 
aggregate blends that were as consistent as possible. Asphalt concrete 
mix designs were conducted for the seven aggregate blends to select the 
optimum asphalt contents. 

Asphalt Cement 

An AC-20 viscosity graded asphalt cement was selected as the asphalt 
material for the natural sand laboratory study. This labstock AC-20 
material is generally considered a medium to hard asphalt cement. An 
AC-20 asphalt cement was selected because of its widespread use across 
the country. The AC-20 material was tested in accordance with ASTM 
D3381 (3) and met the requirements of Table 2 of ASTM D3381. Table 2 
lists the properties of the AC-20 material. 


33 




TABLE 2 

ASPHALT CEMENT PROPERTIES 

(ASTM D3381) 

34 

Test 

Requirements® 

Results 

Viscosity - absolute, 140°F, P 

2000 + 400 

2246 

Viscosity - kinematic, 275°F, Cst 

300 min 

497 

Penetration - 77°F, lOOg, 5 sec, 0.1 mm 

60 min 

80 

Flash Point - Cleveland Open Cup, °F 

450 min 

570 

Solubility in Trichloroethylene - Percent 

99 min 

99.94 

Test on Residue from Thin Film Oven Test 



Percent Weight Loss 


0.21 

Viscosity - 140°F, P 

10,000 max 

5287 

Penetration - 77°F, 100 g, 5 sec, 0.1 mm . 

47 

Ductility - 77°F, 5 cm/min, cm 

50 min 

69.5 


a Table 2 of ASTM D3381 

Natural Sand Materials 

Natural sand material is generally considered to be an aggregate 
that has occurred naturally without any blasting or crushing. A natural 
sand is generally a siliceous material that has a smooth, rounded 
surface and is in the size range between the No. 4 and No. 200 sieves. 
Natural sands can be classified as a fine sand (No. 40 to No. 200), 


medium sand (No. 10 to No. 40) and coarse sand (No. 4 to No. 10). 

Natural sand materials are often used in asphalt concrete mixtures 





35 


because of the low cost and the accessibility of these materials. Two 
locally available natural sand materials were selected for this 
laboratory study. These materials were called mason sand and concrete 
sand. Both of these materials are typical aggregates that are used in 
asphalt concrete mixtures. The mason sand was a medium sand with an 
apparent specific gravity of 2.65 and a water absorption of 0.07 per¬ 
cent. The concrete sand was also a medium sand with an apparent 
specific gravity of 2.64 and a water absorption of 0.20 percent. 

Table 3 lists the aggregate gradations of the mason sand and concrete 
sand. 


TABLE 3 

AGGREGATE GRADATIONS FOR NATURAL SANDS 


U.S. Standard 
Sieve Size 


Mason Sand 
Percent Passing 


Concrete Sand 
Percent Passing 


No. 

8 

100 

100 

No. 

16 

99. 

6 

99.0 

No. 

30 

95. 

6 

80.3 

No. 

50 

47. 

2 

14.0 

No. 

100 

2. 

8 

2.5 


No. 200 


0.5 


1.4 









36 


Limestone Aggregate 

The crushed limestone aggregate used in this study was obtained from 
Vulcan Materials in Alabama. This crushed limestone material is the 
labstock material used in most laboratory research evaluations at WES. 
This material had been separated by a Gilson shaker into various sizes. 
This screening operation processed the material so that the aggregate 
was separated into nine stockpiles, one per sieve size. The limestone 
aggregate had an apparent specific gravity of 2.82 and a water 
absorption of 0.4 and 0.8 percent for the coarse and fine aggregate 
material, respectively. This limestone aggregate had fractured, angular 
faces and a rough surface texture. 

Aggregate Blends 

The laboratory study required that a constant aggregate gradation be 
used throughout the evaluation to decrease the gradation effect on the 
engineering properties of the asphalt mixtures. The 3/4 inch maximum 
aggregate size gradation for high tire pressure applications from 
TM 5-822-8/AFM 88-6 was selected as the target aggregate gradation (10). 
This aggregate gradation was used for all aggregate blends in this 
study. 

Aggregate blends using crushed limestone and various percentages of 
natural sand were blended as closely as possible to the same gradation. 
The aggregate blends contained 0, 10, 20, and 30 percent of each of the 
natural sand materials. As the percentage of natural sand increased, 
especially at 20 and 30 percent levels, the same aggregate gradation was 
not obtainable. As the percentage of natural sand increased, the amount 
of material passing the No. 30 sieve increased. At the 30 percent level 








37 


of natural sand, a definite hump occurred at the No. 30 sieve. The 
aggregate gradations for this laboratory study are listed in Table 4 and 
shown in Figures 11-16. 

As previously mentioned in the literature review (11), a hump in the 
aggregate grading curve that has the sieve sized raised to the 0.45 
power is caused by an excessive amount of natural sand. This hump in 
the aggregate gradation generally occurs between the No. 4 and No. 100 
sieves. Asphalt mixtures that have a hump near the No. 30 sieve are 
most likely to be tender or unstable. The aggregate gradations for this 
laboratory study have been plotted on a chart that has the sieve sizes 
raised to the 0.45 power. These gradations are shown in Figures 17-23. 
It is very evident that as the percentage of natural sand increases, a 
hump at the No. 30 sieve develops. A slight hump is seen at 20 percent 
natural sand while a very distinctive hump is noticed at 30 percent 
natural sand. This indicates that both 20 and 30 percent sand are 


sensitive and tender. 







AGGREGATE GRADATIONS FOR NATURAL SAND LABORATORY STUDY 


o 











cO 

o 

on 

CO 

m 

m 

ro 

00 

ON 



CO 

*H 

CO 

T—1 

VO 

CO 

CM 

m 

m 

o 

m 



O0 

CO 

vO 

m 

Mf 

CO 

H 

H 

o 











CM 

o 

ON 

CO 

m 

m 

VO 

oo 

OO 



i 

o 

• 


• 


• 

. 

. 



w 

rH 

CO 

rH 

VO 

CO 

o 

co 

ON 

H 

in 



oo 

OO 

VO 

in 

Mf 

co 

rH 

rH 




O 












H 

o 

ON 

oo 

in 

m 


m 

vO 

rH 

m 


i 

o 

• 

• 

• 

• 

« 


. 




CO 

t-H 

00 

rH 

VO 

CO 

o 

o 

rH 

CM 

in 

w 

C 



00 

00 

VO 

m 


CO 

CM 

rH 


<D 












H 












CQ 












<U 

33 











•U 

oo 

o 

ON 

00 

m 

in 


<t 

00 

CM 

ON 

«X3 

i 

o 


• 

- 

• 





CO 

H 

oo 

rH 

VO 

co 

o 

00 

CO 

O 

m 

■g 



oo 

00 

VO 

m 


CO 

CM 

rH 


K 
























R 

S 












CM 

o 

ON 

oo 

m 

m 


00 

VO 

uo 

in 


' 

o 


• 

• 

. 

. 

. 


CO 

rH 

00 

H 

VO 

CO 

o 

CO 

CM 

o 

m 




00 

00 

VO 

in 


CO 

CM 

rH 



2C 












H 

100 

ON 

00 

in 

m 

oo 

oo 

CO 

VO 

<± 


CO 

00 

rH 

vo 

CO 

o 

rH 

CO 

rH 

in 




00 

CO 

VO 

m 


CO 

CM 

rH 



O 

i 

O 

o 

ON 

00 

in 

in 

00 

CO 

rH 

<t 

oo 


CO 

rH 

oo 

rH 

VO 

CO 

O 

iH 

CM 

CM 

uo 




oo 

00 

vo 

in 

Mt 

CO 

CM 

rH 














a> 













01 











c 

•U 


VO 

ON 

CO 

O 

OO 

oo 


00 


Q) 

•H 

B 

O 

o 

ON 

1 

OO 


vo 

mT 

CO 

CM 

rH 

vo 

1 

•H 

rH 

CM 

m 

ON 

VO 



in 

oo 

CO 

o 

o 

r-3 


00 

r- 

in 


CO 

CM 

rH 


<D 












aj 





































a> 












N 











C 

CO 

AJ 

cn 

•H 

rC 

X 

X ! 






o 

O 

CO 

a 

o 

O 



vo 

O 

o 

O 

o 

0) 

> 

C 

•H 

p 

•H 

P 

*rl 

Mt 

00 

rH 

CO 

m 

rH 

CM 

u.s. 

0) 

<fr 

CM 

00 

o 

o 

O 

o 

o 

o 

O 

•?H 

CO 

\ 

co 

\ 

rH 

N 

CO 

55 



2; 

:z 


55 










~ q 3/4 1/2 3/8 NO.4 NO.8 NO.16 NO.30 NO.50 NO.IOO NO.200 

SIEVE SIZES 

Figure 11. Aggregate Gradation Curve for S-lM Blend 



Figure 12. Aggregate Gradation Curve for S-2M Blend 




















CLUJCCOLUZH Q.«0«-2O C. LU CC O LU Z h- H,<(/)(0-ZO 



Figure 13. Aggregate Gradation Curve for S-3M Blend 



SIEVE SIZES 

Figure 14. Aggregate Gradation Curve for S-1C blend 





































O2-mm> , 0 Hzmosiimj oz —cow>u nzmonmu 


2 



SIEVE SIZES 

Figure 17. S-0 Gradation Curve Raised to 0.45 Power 



SIEVE SIZES 

Figure 18. S-1M Gradation Curve Raised to 0.45 Power 


















Q-LUCCOUJiH 0.<(0C/5 — ZO Q. lli CC O LU Z I- t<WW-ZO 


3 



SIEVE SIZES 

Figure 19. S-2M Gradation Curve Raised to 0.45 Power 



SIEVE SIZES 

Figure 20. S-3M Gradation Curve Raised to 0.45 Power 















p 

E 

R 

C 

E 

N 

T 

P 

A 

S 

S 

I 

N 

G 



SIEVE SIZES 

Figure 21. S-lC Gradation Curve Raised to 0.45 Power 


P 

E 

R 

C 

E 

N 

T 

P 

A 

S 

S 

I 

N 

G 



Figure 22 


S-2C Gradation Curve Raised to 0.45 Power 

















CLUJCCOUJZH Q.CWCO-ZO 













The Marshall Mix Design procedure, as outlined in Military Standard 
620A (12), was used to determine optimum asphalt contents for the seven 
aggregate blends of this study. Each optimum asphalt content was 
selected at 4 percent voids total mix. The Marshall criteria normally 
used to determine the acceptability of the asphalt content is listed in 
Table 5. The optimum asphalt contents selected from these mix designs 
were used to produce all the specimens for Phase II. 

The Gyratory Testing Machine was used to compact all specimens for 
the mix designs. The gyratory compactive effort used in this study was 
200 psi pressure, 1-degree gyration angle, and 30 revolutions. This 
compaction was equivalent to a 75-blow hand hammer compactive effort 
that is normally used for heavy-duty pavements. 

The Marshall procedure requires that compacted specimens, 4-inches 
in diameter and 2 1/2-inches thick, be tested with the Marshall 
Apparatus which is shown in Figure 24. This procedure is used to 
determine the stability and flow of the asphalt mixture. The stability 
of an asphalt mixture is an indicator of mix strength defined as the 
resistance to deformation under a load. The flow valve is an indicator 
of mix pLasticity measured as the deformation at the maximum load. 

The Marshall procedure requires that a range of asphalt contents be 
evaluated for a given aggregate gradation. Asphalt contents above and 
below the projected optimum asphalt content were evaluated. Data for 
all seven mix designs are listed in Table 6. Each value represents an 
average for three test specimens. The Marshall procedure also requires 
that mixture properties be plotted versus the asphalt content. The 









47 


TABLE 5 

CRITERIA FOR DETERMINING ACCEPTABILITY OF MIXTURE 


Test 

Property 

Heavy-duty 
Pavement 
Requirement (a) 

Marshall stability - lbs 

1800 min 

Unit weight - pcf 

Not used 

Flow - 0.01 inch 

16 max 

Voids total mix - percent 

3 - 5 

Voids filled with asphalt - percent 

70 - 80 


(a) TM 5-822-2/AFM 88-6, Chap 9 
















49 


mixture properties plotted for this study were unit weight, stability, 
flow, voids total mix, voids filled with asphalt, and voids in mineral 
aggregate (VMA). The mix design plots for the seven aggregate blends 
are shown in Figures 25-31. 

The selected optimum asphalt contents are as follows: 


S-0 

5.2 

percent 

S-1M 

4.9 

percent 

S-2M 

4.6 

percent 

S-3M 

4.5 

percent 

S-1C 

4.8 

percent 

S-2C 

4.5 

percent 

S-3C 

4.1 

percent 





ASPHALT CONCRETE MIX DESIGN PROPERTIES 


>> 

H 4-) 

O -r4 X 

CM 

CM 

vO 

o 

co 


CO 

co 

VO 

O 

o 

CM 

4-> r4 <U 

On 

ON 

On 

o 

CM 

VO 

ON 

On 

ON 

o 

o 

m 

(0 T3 

H 43 C 
<0 M 

C5 4J 

CO 

.5 

o ^ 

o 

o 

o 

f-H 

r-H 

f-H 

o 

o 

o 

f-H 

r-H 

f-H 

00 

ON 

rH 

o 

CO 

VO 

00 

00 

ON 

ON 

o 

in 

rH O 

fa • 



|H1 

rH 

r-H 

rH 





f-H 

t-H 

O 


























P 

fH 

rH M 

r—I 


ON 

VO 

ON 

f-H 

VO 


m 


CO 

ON 

-4 43 

VO 

On 

f-H 

r—4 

OO 

VO 

oo 

oo 

in 

ON 

m 

CM 

43 .-4 

o 

CM 

m 


m 

ON 


ON 

ON 

CO 

<r 

oo 

tO ^ 
p 

CM 

CM 

CM 

CM 

CM 

rH 

r-H 

f-H 

f-H 

CM 

CM 

f-H 

m 













rC 

P 

*H P P 













» 3 H C 
•O n) a) 

t—1 

H 

o 

CM 

oo 

ON 


CM 

<t 

o 

f-H 

CO 

•H -O 43 o 

CO 

CM 

o 

VO 

r-H 

CO 

ON 

00 

VO 

CO 

o 

CM 

O <D CX P 

> rH M 0) 

vO 

r- 

00 

00 

ON 

ON 

in 

VO 


oo 

ON 

ON 

<-* <; p. 

•rH ' —' 

Ua 













<D 













C rH P P 

•H cO cO C 

00 

m 

in 

rv. 

rH 

ON 

o 



ON 

r-H 

ON 

u to a> 


• 

• 

• 

• 

• 


• 

• 

• 

• 

. 

0) 0> 0) o 

VO 

VO 

vO 

VO 


r-* 

vO 

m 

in 

m 

vO 

VO 

T3 C H H 

f —< 

»—1 

f-H 

rH 

f-H 

f-H 

r-H 

r-H 

f-H 

f-H 

r*H 

f-H 

•H -r4 b0 <U 

o s; m o. 

> < 














O) rH 
X) 

•H U 

o o 
> H 


c 


WOO’ 
< O (X 


<y 

p 

co *d 
bO C 
0) <u 
p »—f 
bo cq 

to 

< 


/-N 

P 

c 

CM 

vO 

CO 

co 


rH 

m 

o 


r- 

vo 

co 

X 0) 


• 

• 

• 





• 


. 

• 

•H O 

VO 


CO 

CM 

rH 


vO 

in 

co 

CM 



S P 

Q) 

tx 

P i 

t-H 

VO 


00 

rH 

CO 

CM 

CO 

CM 

n- 

CM 

CO 

C ^ 

W) o 

CO 

vj- 

m 

m 

vO 

in 

co 


in 

m 

VO 

m 

*H (X 

in 

in 

m 

m 

in 

m 

m 

in 

m 

m 

m 

m 

<D 

t-H 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

/-N 

p p 

c c 

m 

o 

in 

O 

in 

o 

o 

in 

o 

m 

O 

in 

a) <d 













P u 

C H 


m 

in 

VO 

vO 




m 

m 

vO 

VO 


o 

I 

m 















TABLE 6 (continued) 


^ J-) 

o -w x 

4J r-l 0) 

M Tl t) 

u ja c 

I>1 « H 
O 4-> 

CO 


x 

4J 

•H 4J 4J 


•H X> J3 O 

o a) cx n 

> rH W <D 
I-I < a 


<y /“s 

Ch u u 

•rl 14 « C 

>-* bo a) 
in a) <u o 
x) C n >-i 

•H «H b0 <U 
O S U> P. 
> < 


x) n) x a) 
.h .u *h o 
o o g n 


W U (J 
rH C C 
cfl 0) 0) 
X O 
o. c n 
to o <u 
< u a. 


co 

CO 

o 


CO 

o 

CO 

vO 

vo 

o 


O' 

O' 

o 

O 

*H• 

o 

O' 

O' 

On 

o 

H 

o 

o 

T — \ 

rH 

rH 

CM 

O 

O 

O 

H 

H 


in 

ON 





m 

vO 

00 

H 

CO 

r-* 

o 

ON 

CM 

VO 

r^ 

O 

rH 

vO 

CM 

<r 

m 

m 

H 

CM 

CN 

rH 

»—1 

rH 

H 

i —i 


r~- o\ c«-> 

O rH o 

ro 


CM 

H 

CM 

CM 

ON 

O 

rH 

CM 


00 

o 

r- 

m 

in 

m 

in 

m 


in 

in 



in 

in 

H 

t—i 

H 

H 

rH 

rH 

rH 

rH 

»—1 

T—4 

T—l 

H 



CM 

O 

in 

in 

ON 


o 

00 


CO 

CM 

rH 

rH 

vO 

in 


CM 


4J /~s 
u r*, (i-i 

VO 


CM 

ON 


CO 

ON 

CM 

CM 

On 

CO 

too 

CO 


m 

m 

m 


f—1 

CO 



in 

C *r4 Oi 

m 

in 

in 

in 

m 

m 

m 

m 

m 

m 

m 

P 0) w 

13 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 

rH 


to in vo vo 













TABLE 6 (continued) 
















TABLE 6 (continued) 


o -P x 

•U r-l a> 
(fl "P "O 

p & C 
>> « w 
CJ p 

GO 


•H 








rH 

w 

o 


rH 

CO 

in 

*H 

•H 

•a 

CM 


On 

VO 

co 


£> 

t—f 

co 


-vf 

VO 

oo 

m 

cd 

4J 

V-' 

H 

i — < 

rH 

rH 

rH 

rH 


X, 

P /-v 
•P 4-1 P 
M tE ,P C 
X) (0 <1) 
•ft X) ,C o 

o o) p,n 

> rH 01 <u 

*-< < a. 


Q) ✓-’s. 

C rl W W 
•h n) tfl C 

P bo ai 
oi a) <u o 
t) G u u 
•p .p t»o ai 
os «a 
> < — 


W 

rH 

c o 

vO 

CM 

o 

T> 

nj X 

0) 




•H 

p -p 

u r- 

uo 


CO 

O 

O X 





> 

H 

a> 





P 


OV 

CO 

VO 

m 

CM 

p .c <p 

• 





•P bO O 

CM 

CO 

Mt 

in 

vO 

VO 

G -P (X 

in 

in 

m 

m 

uo 

in 

3 0'-' 

rH 

?H 

»H 

» —< 

rH 

rH 


u u u 

r-f G G 

al <D <v 
£. P O 
cx G P 

01 O 4) 
< U (X 
















o * 



Figure 26. Mix Design Plots for S-1M Blend 

















so -I* 4 



































»<0 -» f 
















































61 


Mixture Properties at Optimum Asphalt Content 

Table 7 lists a summary of the mix design properties at the optimum 
asphalt content for each aggregate blend. Several observations and 
trends were observed from the mixture properties. The optimum asphalt 
content for each natural sand material decreased as the percentage of 
natural sand material increased. The optimum asphalt content for the 
mason sand blends decreased from 5.2 percent at 0 percent sand to 
4.5 percent at 30 percent sand. The optimum asphalt content for the 
concrete sand blends also decreased from 5.2 percent at 0 percent sand 
to 4.1 percent at 30 percent sand. Figures 32-33 show the optimum 
asphalt content versus percent sand in mixture. 

The stability value for the aggregate blends at the optimum asphalt 
content decreased as the percentage of natural sand increased. The 
stability value for the mason sand blends decreased from 2395 lbs at 
0 percent sand to 1570 lbs at 30 percent sand. The stability value for 
the concrete sand blends decreased from 2395 lbs at 0 percent sand to 
1550 lbs at 30 percent sand, a reduction in stability of approximately 
35 percent. The stability values versus percent sand in mixture are 
shown in Figures 34-35. Another trend that was observed in the 
selection of optimum asphalt contents was a decrease in voids in mineral 
aggregate (VMA) as the percentage of natural sand increased. The VMA 
value for the mason sand blends decreased from 16.4 percent at 0 percent 
sand to 14.7 percent at 30 percent sand. The VMA value for the concrete 
sand blends also decreased from 16.4 percent at 0 percent sand to 
13.8 percent at 30 percent sand. The VMA values versus percent sand in 
mixture are shown in Figures 36-37. 







SUMMARY OF MIX DESIGN PROPERTIES AT OPTIMUM ASPHALT CONTENT 


I 



X 










4-> 









CO 

i-4 

4-) 4-> 








•o 

U 

f-i c 

o 

o 

o 

o 

o 

o 

o 

**-4 


cd a) 

• 

• 

• 

• 

• 

• 

• 

o 

T3 

o 

m 

in 

co 

co 



CM 

> 

O 

P- ^ 

r- 



r*» 





H 

W <D 









»— 4 

C cx 









T-l 

N-' 









C*-< 











<13 x-s 








C 

r—4 

4-> 4J 








*r-l 

cfl 

cd C 

<r 

r-* 

o 

r- 

m 

CO 

00 


V4 

bo a> 

• 

• 

• 

• 

• 

• 

• 

W 

<D 

a> o 

vO 

m 

in 


m 


co 

T3 

C 

P u 

H 

r— • 

r—4 

r—4 

t—4 

H 

r-t 


•t-4 

bO <D 








o 

SJ 

bO CX 








> 


< w 




















P 








w 

r—4 

C 








X> 

cd 

X <D 

O 

o 

o 

o 

o 

o 

O 

•r-4 

4-1 

•f-l O 

• 

• 

• 

• 

• 



o 

O 

S u 


<r 





Ml - 

> 

H 

<v 










Ou 



















/~s 








c 








Flow 
01 i 

ON 

CO 

ON 

On 

00 

oo 

00 

o 
















4-> 

•r4 

m 

o 

o 

o 

o 

o 

o 

r-t 10 

ON 

CO 

oo 

r- 

o 

o 

in 

-H XI 

co 

r—1 

00 

in 

CM 

On 

in 

x> » — t 
<d ' 

4-1 

co 

CM 

CM 

i—4 

r—1 

CM 

H 

i—i 

4-1 

ON 

o 

VO 

CM 

o 

CO 

ON 

4J X 
•H b0 O 


in 



in 

m 


c -r-i a. 

m 

in 

in 

m 

m 

m 

in 

DO)'— 


» — i 

»—4 

r—4 

r—4 

r—1 

T—l 

3 








6 4->4-> 

3h c 








B id a> 

•H JZ U 

CM 

On 

vO 

m 

oo 

m 

i—4 

w p. c 
o. <n o 
o < u 

in 

-j- 





<f 

a) 

AJ 

td XJ 


s 

£ 

£ 

o 

o 

CJ 

W) c 

O 

iH 

CM 

rO 

r— 4 

CM 

CO 

<U 0) 

>-4 r—4 

60 CO 

60 

■ 

in 

i 

co 

< 

co 

1 

CO 

i 

CO 

CO 

i 

CO 

< 






















o 


10 


2 


Y 


Percent Sand In Mixture 


30 


Figure 32. Optimum Asphalt Content Versus Percent Mason Sand 


























><:< 


Percent Sand in Mixture 

Figure 36. Voids in Mineral Aggregate Versus Percent Mason Sand 


% 



0 10 20 30 

Percent Sand in Mixture 


Figure 37. Voids in Mineral Aggregate Versus Percent Concrete Sand 










CHAPTER V 


PHASE II - PRESENTATION AND ANALYSIS OF DATA 


This chapter presents and discusses the results of the laboratory 
testing involved in Phase II of this laboratory study. Phase II testing 
was developed to evaluate the effects of natural sands on asphalt 
concrete mixtures using state-of-the-art testing equipment. Forty 
asphalt concrete specimens were produced for each aggregate blend at the 
optimum asphalt content determined in Phase I. Each specimen was com¬ 
pacted with the GTM. The Marshall stability, flow and voids properties 
were determined for each aggregate blend. The indirect tensile, resil¬ 
ient modulus, and unconfined creep-rebound tests were also conducted to 
determine the strength characteristics of the various mixtures. 

The purpose of this study was to determine the effect of natural 
sands on asphalt mixtures and to determine allowable limits for the 
natural sand content. The general approach used to analyze the test 
results involved a direct comparison of test values and a graphical 
analysis. The scope of this laboratory study allowed a direct 
comparison of test values because the main variable was the amount or 
percentage of natural sand in the mixture. Since the number of 
variables was limited, the comparison of these results for each test 
procedure was considered to be an excellent means of analyzing these 
mixtures. 


66 










67 


Graphical analyses were conducted for practically all test results 
for this study. A large number of graphs were produced to allow a 
visual interpretation of the data. Graphical analyses generally 
demonstrate trends and tendencies and exhibit test variable relation¬ 
ships. The graphs produced in this study supported the expected 
findings and helped define certain relationships and trends. 

Marshall Mix Properties 

Phase II of the natural sand laboratory evaluation required that the 
standard Marshall mix properties be determined at the optimum asphalt 
content so these test values could be analyzed with the more modern, 
sophisticated test procedures. A summary of the Marshall mix properties 
for Phase II is presented in Table 8. The test results presented for 
the mix properties, unit weight, voids total mix, voids in mineral 
aggregate, voids filled with asphalt, and the gyratory elasto-plastic 
index (GEPI) are an average of 40 specimens. The stability and flow 
test results are an average of three to nine specimens. 

The optimum asphalt contents that were selected in Phase I were 
based on mixtures having 4 percent total voids. The percent voids total 
mix for the specimens produced in Phase II varied slightly from the 
target value. The average percent voids total mix maximum variance from 
the target value was 0.3 percent for the S-0 aggregate blend. The 
remaining average values had less than an 0.2 percent variance. These 
variances in percent voids total mix are not considered to be signifi¬ 


cant and should not have an effect on the test results. 






SUMMARY OF MIX PROPERTIES AT OPTIMUM ASPHALT CONTENT 



68 











69 


The unit weight or density values did not vary significantly as the 
percentage of natural sand increased. The unit weight of all the 
crushed limestone mixture (S-0) was 154.6 pounds per cubic foot (pcf). 
The unit weight values for the mixtures containing natural sand did not 
vary significantly from the S-0 blend. The maximum unit weight value 
was 155.3 pcf for the S-1M and S-2C blends and the minimum unit weight 
value vas 154.0 pcf for the S-3M blend. The difference in unit weight 
values from the S-0 blend is less than 1 pcf and was consideied to be 
insignificant. 

The voids filled with asphalt values indicated a general trend that 
these values decreased as the percentage of natural sand increased. The 
voids filled values for the S-0 blend was 74.4 percent. The test 
results showed a small variance at 10 percent natural sand, but a 
larger, more significant variance at 20 and 30 percent natural sand. 

The voids filled with asphalt value was 75.6 percent for the S-1M blend 
and 73.6 percent for the S-1C blend. The voids filled value decreased 
to 72.3 percent for the S-3M blend and 70.1 per-cent for the S-3C blend. 

The voids in mineral aggregate (ViiA) test results also decreased as 
the percentage of natural sand increased. The VMA value for the 
S-0 blend was 16.7 percent. The asphalt mixtures containing natural 
sand progressively decreased from this value. The average value 
was 15.6 percent for 10 percent natural sand, 14.8 percent for 20 per¬ 
cent natural sand, and 14.4 percent for 30 percent natural sand. 

Figures 38-39 show the VMA values versus percent natural sand in 
mixture. This relationship of decreasing VMA values with increasing 
percentages of natural sand is supported in both Phases I and II. This 







Figure 38. VMA Versus Percent Mason Sand 



0 10 20 

Percent Sand in Mixture 

Figure 39. VMA Versus Percent Concrete Sand 


30 











71 


reduction in VMA values indicated the potential for less stable asphalt 
mixtures and unsatisfactory field performance (15). 

The Marshall stability test results indicate there is a direct 
relationship between stability and the percentage of natural sand. As 
the percentage of natural sand increases in an asphalt concrete mixture, 
the stability or resistance to deformation decreases significantly. The 
stability values for each aggregate blend are listed in Table 9. The 
stability value for the crushed aggregate mixture (S-0) had an average 
stability of 2393 lbs. This value is well above the 1800 lbs minimum 
requirement for heavy duty pavements. The decrease in stability values 
was minor for 10 percent natural sand, approximately 4.9 percent. The 
decrease was more pronounced at the 20 and 30 percent natural sand 
contents, 20.5 percent and 32.9 percent, respectively. At the 30 per¬ 
cent level of natural sand, the stability values had decreased to 
approximately 1600 lbs which is below the minimum requirement and not 
acceptable for heavy duty pavements. Table 10 lists the summary of 
Marshall stability values and Figures 40-41 show these values versus the 
percent sand in mixture. 

The Marshall flow values did not indicate a significant relationship 
between flow values and percent natural sand. A larger effect on flow 
values was caused by the type of natural sand instead of percentage of 
sand. The mason sand had little effect on the flow of the mixtures; all 
mixtures had a flow of 9. The concrete sand caused a larger change; a 
flow value of 7 at 30 percent natural sand. 





72 


TABLE 9 

MARSHALL STABILITY AND FLOW RESULTS 


Marshall 

Aggregate Stability Flow 

Blend (lbs) (0.01 in) 


S-0 2133 10 

2183 10 

2218 10 

2756 9 

2617 10 

2450 10 

S-1M 2080 9 

2288 7 

2664 7 

2354 8 

2432 8 

2496 8 

S-2M 1924 8 

1832 9 

1742 9 

1832 9 

1786 8 


1786 


9 










73 


TABLE 9 (continued) 

Marshall 

Aggregate Stability Flow 

Biend (lbs) (0.01 in) 

S-3M 1483 9 

1578 8 

1526 9 

1768 10 

1786 10 

1638 10 

S-1C 2098 8 

1950 8 

2270 9 

2054 8 

2508 8 

2184 8 

1936 8 

2328 8 

S-2C 1976 7 

1964 7 


1986 


7 




74 


TABLE 9 (continued) 


Aggregate 

Blend 

Marshall 

Stability 

(lbs) 

Flow 

(0.01 in) 

S-3C 

1526 

7 


1578 

7 


1392 

7 


1924 

7 


1860 

7 


1578 

7 


1482 

7 


1508 

7 


1378 

7 












75 


TABLE 10 

SUMMARY OF MARSHALL STABILITY VALUES 


Percent 

Natural 

Sand 

Type of 

Sand 


Marshall 

Stability 

(lbs) 

Percent 

Decrease 

0 

Crushed 


2393 


10 

Mason 


2386 

0.3 


Concrete 


2166 

9.5 


Average for 

10% 

Sand 2276 

4.9 

20 

Mason 


1817 

24.1 


Concrete 


1986 

17.0 


Average for 

20% 

Sand 1902 

20.5 

30 

Mason 


1630 

31.9 


Concrete 


1581 

33.9 


Average for 30% Sand 1606 


32.9 









76 



Figure 40. Marshall Stability Versus Percent Mason Sand 



10 20 30 

Percent Sand in Mixture 













77 


Indirect Tensile 

The indirect tensile test was conducted to determine the fundamental 
tensile strength properties of the asphalt concrete mixtures. This test 
was conducted on a minimum cf three specimens for each of the seven 
aggregate blends. The indirect tensile test was conducted at two test 
temperatures, 77°F and 104°F. These test temperatures were chosen 
because most pavement deformation occurs at higher temperatures. The 
results of the indirect tensile test are presented in Table 11. 

Tensile strength values are usually dependent on the type of binder 
or asphalt cement material and the temperature of the testing. The test 
results of this study indicate that the test temperature had a signifi¬ 
cant effect on the tensile strength values. The tensile strength values 
at 77°F are approximately three times greater than the tensile strength 
values at 104°F. A summary of tensile strength values at 77°F and 104°F 
are presented in Tables 12-13. 

The tensile strength values were also affected by the percentage of 
natural sand in the mixture. At 77°F, the tensile strength of the all 
crushed limestone mixture (S-0) was 147.0 psi. The tensile strength 
values for the mixtures containing natural sand decreased as the per¬ 
centage of natural sand increased. The average tensile strength value 
was 125.7 psi for 10 percent natural sand, 118.7- psi for 20 percent 
natural sand, and 116.3 psi for 30 percent natural sand. The reduction 
in tensile strength at 30 percent natural sand was approximately 
20.9 percent. The actual tensile strength decreased for the mason s.nd 
specimens was 28.9 percent. 






78 


Figure 42 shows the indirect tensile strength values at 77°F versus the 
percent natural sand in mixture. 

The indirect tensile strength values at 104°F were also affected by 
an increase in natural sand materials. The indirect tensile values 
decreased significantly as the percentage of natural sand increased. 

The indirect tensile strength for the S-0 blend was 50.1 psi. The 
average tensile strength value was 42.9 psi for 10 percent natural sand, 
41.0 psi for 20 percent natural sand, and 37.9 psi for 30 percent 
natural sand. The decrease in tensile strength at the 30 percent 
natural sand content was 24.4 percent. Figure 43 shows the indirect 
tensile strength values at 104°F versus the percent natural sand in the 
mixture. 

Resilient Modulus 

The resilient modulus test was conducted to evaluate the relative 
quality of the asphalt concrete mixtures. This test was conducted on a 
minimum of three specimens for each of the seven aggregate blends. 

Since this test was considered to be a nondestructive test, duplicate 
tests were conducted on each specimen. The resilient modulus test was 
also conducted at two test temperatures, 77°F and 104°F. The results of 
the resilient modulus test are presented in Table 14. 

The resilient modulus value of an asphalt concrete mixture is 
generally dependent on the type of asphalt cement, aggregate gradation, 
and the shape and texture of the aggregate. Since this laboratory study 
used the same asphalt cement and primarily the same aggregate gradation, 
the variation in aggregate shape and texture would be analyzed. 










79 


TABLE 11 

INDIRECT TENSILE TEST RESULTS 


Aggregate 

Blend 

Temperature 
(degrees F) 

Thickness 

(inches) 

Vertical 

Load 

(pounds) 

Tensile 

Strength 

(psi) 

S-0 

77 

2.514 

2317.2 

146.7 


77 

2.515 

2432.8 

154.0 


77 

2.524 

2204.3 

139.0 


77 

2.469 

2379.0 

153.4 


77 

2.504 

2231.2 

141.8 


104 

2.531 

811.8 

51.1 


104 

2.500 

814.5 

51.9 


104 

2.492 

787.6 

50.3 


104 

2.505 

752.7 

47.8 


104 

2.502 

776.9 

49.4 

S-1M 

77 

2.462 

2024.2 

130.9 


77 

2.466 

1922.1 

124.1 


77 

2.507 

1954.3 

124.1 


77 

2.503 

1908.6 

121.4 


77 

2.477 

2013.4 

129.4 











80 


TABLE 11 (continued) 


Aggregate 

Blend 

Temperature 
(degrees F) 

Thickness 
(inches) 

Vertical 

Load 

(pounds) 

Tensile 

Strength 

(psi) 

S-lM 

104 

2.449 

750.0 

48.7 


104 

2.480 

704.3 

45.2 


104 

2.495 

728.5 

46.5 


104 

2.483 

701.6 

45.0 


104 

2.479 

707.0 

45.4 

S-2M 

77 

2.503 

1798.4 

114.4 


77 

2.493 

1828.0 

116.7 


77 

2.504 

1743.6 

110.8 


77 

2.494 

1771.3 

113.0 


77 

2.503 

1710.6 

108.8 


104 

2.481 

626.3 

40.2 


104 

2.503 

604.8 

38.5 


104 

2.510 

611.7 

38.8 


104 

2.487 

729.8 

46.7 


104 

2.493 

681.9 

43.5 

S-3M 

77 

2.484 

1619.4 

103.8 


77 

2.511 

1612.9 

102.2 


77 

2.497 

1618.3 

103.2 


77 

2.503 

1707.5 

108.6 


77 


2.504 


1646.2 


104.6 









81 


TABLE 11 (continued) 


Aggregate 

Blend 

Temperature 
(degrees F) 

Thickness 

(inches) 

Vertical 

Load 

(pounds) 

Tensile 

Strength 

(psi) 

S-3M 

104 

2.501 

611.8 

38.9 


104 

2.521 

622.6 

39.3 


104 

2.489 

557.0 

35.6 


104 

2.498 

548.4 

34.9 


104 

2.497 

665.6 

42.4 

S-1C 

77 

2.490 

1994.6 

127.5 


77 

2.492 

2010.8 

128.4 


77 

2.484 

2008.6 

128.7 


77 

2.492 

2034.4 

129.9 


77 

2.509 

1773.1 

112.5 


104 

2.496 

599.5 

38.2 


104 

2.498 

588.7 

37.5 


104 

2.481 

646.2 

41.5 


104 

2.479 

623.7 

40.0 


104 

2.453 

625.8 

40.6 

S-2C 

77 

2.491 

1969.9 

125.9 


77 

2.443 

1872.0 

122.0 


77 

2.453 

1941.9 

126.0 






82 


TABLE 11 (continued) 


Aggregate 

Blend 

Temperature 
(degrees F) 

Thickness 

(inches) 

Vertical 

Load 

(pounds) 

Tensile 

Strength 

(psi) 

S-2C 

104 

2.495 

619.4 

39.9 


104 

2.458 

641.9 

41.2 


104 

2.488 

673.4 

39.6 


104 

2.479 

608.5 

41.2 

S-3C 

77 

2.501 

2167.0 

137.9 


77 

2.493 

2096.8 

133.9 


77 

2.496 

1955.9 

124.7 


77 

2.518 

1855.9 

117.3 


77 

2.483 

1973.1 

126.5 


104 

2.500 

673.4 

42.9 


104 

2.503 

608.5 

38.7 


104 

2.495 

554.8 

35.4 


104 

2.476 

591.4 

38.0 


104 


2.481 


511.8 


32.8 








83 


TABLE 12 

SUMMARY OF INDIRECT TENSILE TEST AT 77°F 


Percent 

Natural 

Sand 

Type of 

Sand 


Tensile 

Strength 

(psi) 

Percent 

Decrease 

0 

Crushed 


147.0 


10 

Mason 


126.0 

14.3 


Concrete 


125.4 

14.7 


Average for 

10% 

Sand 125.7 

14.5 

20 

Mason 


112.7 

23.3 


Concrete 


124.6 

15.2 


Average for 

20% 

Sand 118.7 

19.3 

30 

Mason 


104.5 

28.9 


Concrete 


128.1 

12.9 


Average for 

30% 

Sand 116.3 

20.9 







84 


TABLE 13 

SUMMARY OF INDIRECT TENSILE TEST AT 104°F 


Percent Tensile 

Natural Type of Strength Percent 


Sand 

Sand 



(psi) 

Decrease 

0 

Crushed 



50.1 


10 

Mason 



46.2 

7.8 


Concrete 



39.6 

21.0 


Average for 

10% 

Sand 

42.9 

14.4 

20 

Mason 



41.5 

17.2 


Concrete 



40.5 

19.2 


Average for 

20% 

Sand 

41.0 

18.2 

30 

Mason 



38.2 

23.8 


Concrete 



37.6 

25.0 


Average for 30% Sand 37.9 


24.4 












Percent Sand in Mixture 

Figure 43. Indirect Tensile Strength Values at 104°F 







86 


However, Che results from Che resilient modulus tests were inconsistent 
and showed no conclusive trends. 

The inconsistency of the data was very evident when duplicate test 
values from the same specimen were evaluated. Two-thirds of the 
specimens tested had results that varied from the initial test value by 
more than + 20 percent.’ The vast majority of the second test values had 
increased when compared to the initial test value. The variation in 
test values ranged from a 50 percent decrease to a 200 percent increase. 
Based on this significant variation in test results, only the initial 
resilient modulus values were analyzed. Two initial test values were 
also eliminated because these values were approximately five times 
greater than the other two specimens at the same asphalt content and 
gradation. These test values were approximately twc million psi, not 
typical values for an asphalt concrete mixture at 77°F. 

The resilient modulus values that were analyzed indicated that the 
test temperature and the amount of natural sand did effect the resilient 
modulus values. The resilient modulus values at 77°F were three to five 
times greater than the resilient modulus values at 104°F. The resilient 
modulus values also decreased as the percentage of natural sand 
increased, but the values were inconsistent. A summary of the resilient 
modulus values at 77°F and 104°F are presented in Table 15. 

The resilient modulus values at 77°F indicated the type of natural 
sand had some effect on the resilient modulus value. The resilient 
modulus value for the crushed limestone mixture (S-0) was 589,192 psi. 
The resilient modulus value was 547,194 psi for 10 percent mason sand, 
465,744 psi for 20 percent mason sand, and 390,828 psi for 30 percent 





87 


mason sand. The resilient modulus value was 492,214 psi for 10 concrete 
sand, 423,814 psi for 20 percent concrete sand, and 579,898 psi for 
30 percent concrete sand. The various amounts of natural sand did not 
develop a true relationship for the resilient modulus value at 77°F. 
Figure 44 presents the resilient modulus values at 77°F versus the 
percent sand in mixture. 

The resilient modulus values at 104°F also indicated an inconsistent 
relationship between the resilient modulus value and the percentage of 
natural sand in the asphalt concrete mixture. The resilient modulus 
value for the S-0 blend was 190,354 psi. The resilient modulus value 
was 164,722 psi for 10 percent mason sand, 199,522 psi for 20 percent 
mason sand, and 147,414 psi for 30 percent mason sand. The resilient 
modulus value was 99,412 psi for 10 percent concrete sand, 126,833 psi 
for 20 percent concrete sand, and 140,431 psi for 30 percent concrete 
sand. These resilient modulus values are varied enough to be considered 
inconsistent. Figure 45 presents the resilient modulus values at 104°F 
versus the percent natural sand in mixture. 

















TABLE 14 (continued) 


89 


p 


c 

to 

<}■ 


r- 

On 

CNI 

H 

VO 

VO 

co 

CO 

r>* 

00 

a> 

d _ 

On 

o 

VO 

O 

CNJ 

rH 


o 

o 

oo 

oo 

oo 

*rl 

,_l 

tH 

On 

co 

00 


CNJ 


CNJ 


VO 

o» 

CO 

» — t 


«■ 

«*■ 

- 

- 

- 

- 

- 

- 

- 

- 



•H 

•d 2 



On 

<1- 

VO 

cn 


o 


CO 

00 

VO 

W 

o £ 

VO 

ro 

00 

00 

oo 


oo 

*4- 

i—i 

oo 

oo 

f-"- 

o 

s 

m 

^4 

r—1 

in 

CO 


»—i 

»—1 

»—i 

r— 1 

«—1 

rH 


oi 


■3 £ ~ 

tfl o . 


P -H o 


c 

P rP 

ON 

CO 

oo 

On 

VO 

CO 

CO 

r*-* 

o 

CO 

CO 

CO 

O 

d 

in 

T—1 


m 

o 

*—< 

rH 

vO 

CNJ 

«H 



N 

6 u) 

r—1 

CNI 


i—< 

1 —< 

CM 

CNJ 

CN 

CO 

CM 

CNI 

CNI 


U 0) 

• 

• 

• 


• 

• 

• 

• 

• 

. 


. 

U 

c 

o fi. 

4-1 o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

O 

o 

o 


51 «) C 
a *p 


rt 

(/) 

ON 


ON 

<T 



ON 

CO 

CO 

ON 

On 

CO 

o d 

X5 

in 

CO 

m 

C0 

ro 

CO 

CM 

m 

m 

CM 

CM 

in 

•H d 

q 

in 

VO 

in 

VO 

VO 

vo 

i-H 

o 

o 

«H 

i — l 

o 

n o 

3 





• 

• 


• 





U rJ 

o 

vo 

n* 

vo 



r^ 

VO 

in 

m 

VO 

vo 

in 

a) 

a- 

CO 

CO 

CO 

CO 

CO 

CO 


•H 

i-H 

i — 1 

rH 

rH 


> 


C w 

O 0) 
•p <u 

p p 

nj bO 
p a) 
o "d 
eJ ^ 


o 


o o o o o o o 

CTv O' OS O' 


o o 

O' 


o 


o 

O' 


U) /•"N 

1/5 05 
<D a) 


r** 

CO 

CO 


n* 

in 

m 

co 

CO 

ON 

ON 

C .d 

o 

o 

o 

o 


fN 

On 

ON 

oo 

OO 

r- 


a: o 

m 

m 

in 

in 

pt 

pf 

<4 


<4 

>4 

Mt 

pt 

o c 

H H 

CM 

CM 

CM 

CM 

CM 

CN 

CM 

CM 

CM 

CM 

CM 

CM 




fl) 

P 


3 

p 


n) <u 
p p 
a) bO 
(X a) 
6 -d 

<U s_ ' 

H 



<U 

P 

d "d 
bO C 
<U 0 ) 
P rP 
bO OQ 
bO 
< 


CO 






TABLE 14 (continued) 


•U 


c 

W 

00 

rv» 

m 

oo 

00 

1—1 

ON 

o 

m 

ON 

CO 

00 

<U 

3 _ 

co 

co 

r—1 

co 

IV 

on 

in 

l£> 

ON 

CM 

i—1 


*r— ( 


CO 

co 

00 

00 

o 

KD 



CNJ 

CO 

OO 

CO 

f—1 

a t; 

- 

« 

«* 

*. 

- 

- 



- 




*H 

•o “ 



r—1 

CM 

oo 

1—1 

CO 

00 

o 

CM 

ON 

CM 

W 

o ,°r 

ON 

on 

H 

CO 

CO 

CO 

ON 

On 

o 

r-* 

ON 

VO 



Mt 


m 

CO 

CO 

CO 

r—1 

t—♦ 

CM 


i—1 

H 


cfl O . 


« o 


c 

4-> i-l 



<* 

1—1 

00 

H 

T—1 

i—i 

1—1 

CM 

1—1 

o 

«d 

T—1 

H 

1—1 

CM 

VO 

CM 

VO 

vO 

VO 

00 

vO 

N 

B w 

CM 

CM 

CM 

CO 

CM 

CO 

1—1 

rH 

1—1 


1-1 


U Q) 






• 






u 

o x: 

o 

o 

o 

O 

O 

o 

O 

o 

o 

o 

O 


O <4-1 O 

X a) C 


Q -H 



w 

o 

o 

VO 

o 

m 

o 

CO 

CO 

CO 

00 

CO 

00 

a x> 

X> 

1—1 

i — 1 

OO 

1—1 

CO 

r—1 

o 

o 

o 

rv« 

O 

n- 

•H <Tl 

c 

o 

o 

o 

o 

On 

o 

ON 

ON 

On 

On 

ON 

ON 

4-> O 

3 





. 

. 




^ -4 

o 

CO 

CO 


CO 

i—l 

CO 

CM 

CM 

CM 

CO 

CM 

CO 

o 

a 



vt 




«—1 

»—l 

T—1 

»—1 

r—1 

r-i 


c 

o 


tn 
<u 
ai 
U H 
(fl HO 
4J <1) 

o -a 


oooooooooooo 

O' on o' on ct> as 


w 


V) 

W 

<D 

a 

c 

X 


o 

O 

c 

•r~l 

•rl 

X 


H 




<fr 


Mf 

co 

co 

o 

O 



co 

CO 

o 

O 

ON 

ON 

O 

O 

i—i 

T—1 

oo 

oo 

ON 

On 

m 

in 



in 

in 

in 

in 



<t 

Mt 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 


<u 

Vj \ 

3 w 

4-> <D 

<ti a> 

0) 00 
PL (D 

e x) 

<D w 

H 


r-. 


r>. r>. r^ 

r- 


O 


<r 

o 


o 


o 


o 


o 


0) 

u 

03 X3 
bO C 
a) a) 

V-J ?—< 
bO CQ 
bO 
< 


S 

CM 

on 


90 








TABLE 14 (continued) 


91 


u 












e 

w 


m 

o 

VO 

CNJ 

CNJ 

vO 

CO 


o 

<u 

3 


<1* 


rH 

CNJ 

CNJ 

tH 

O 

vo 

o 

*H 
» —1 
H 

rH 

3 

X) 

H 

« 

CO 

oo 

ON 

VO 

CNJ 

uo 

oo 

o 

m 

r- 

O 

cnj 

ON 

cn 

o 


00 

ON 

OO 

H 

ON 

rH 

m 

in 

CNJ 

0) 

s 


CO 

CO 

CO 

UO 

CO 

UO 

CNJ 

«4- 

fH 


co o 

UO vO 

VO VO 

VO VO 


c.T' 

o • 

*H O 
•U H 
CTJ 

W 

a) 
O rC 
0 4-10 
X <1) C 
a *h 


<TJ 
4J 

c 

o 

N S 






o 



CO 


o 

f J 

o 

rH 

rH 

tH 

rH 

VO 

rH 

o 

m 

rH 

vo 

CNJ 

vO 

CNJ 

CNJ 

CNJ 

CNI 

»H 

CNJ 

»H 

o 

CNJ 

rH 

CO 

rH 

o 

o 

O 

o 

O 

O 

o 

o 

O 

o 

o 

O 


O T3 T3 


0) 


c 

o 

a- 


o 

oo 

CO 

CO 

00 

CO 

CNI 

r^ 

CO 

o 

CO 

CO 

o 

CO 

in 


CO 


CO 

CO 


CO 


vo 

CO 


CO 

CO 

<!■ 

CO 

O 

ON 

CO 

CO 

CO 

CO 

CO 

CO 

rH 



CNJ 

in 

o 


CNI 

CNI 

CNJ 

uo 

o 

UO 


vo 

r- 

o 

00 

o 


C 

o 


w 
o 
<u 

4-> V4 
cd bO 
4~> 0) 
O *0 

a: — 


o 

on 


o 

on 


o 

on 


o 

ON 


o 

on 


o 

ON 


W 

in m 


0) <D 

r>. 


CO 

CO 


>3- 

ON 

ON 

oo 

00 

r- 

r— 

V-* rH 

ON 

ON 

o 

o 

o 

o 

oo 

oo 

ON 

ON 

O' 

ON 

A2 O 

O C 



uo 

m 

uo 

uo 







•H H 

CNJ 

CNJ 

CNI 

CNI 

CNJ 

CNJ 

CNI 

CNJ 

CNJ 

CNJ 

CNI 

CNI 


H 


<i> 

V-J /"-v 

3 w 
•u a) 
n; 

u u 
a> no 
a <u 

£ T3 

(D w 

H 






<t 

o 




<f 

o 


<u 

u 

<0 T3 
bO C 
<D <D 
V* rH 

bO CQ 
bO 
< 


co 

i 

CO 


166,600 








TABLE 14 (continued) 



104 







TABLE 14 (continued) 


93 


4-1 

C 01 

cO 

CM 

•st 

OO 

o 

On 


co 

5 3 ^ 

fH 

<* 

i—1 

CM 

VO 

t—1 


tH 

•rl rH . 

CO 

NO 

co 

VO 

in 

VO 

ON 

rH 


rH 

O 

O 

vo 

*H 

o 

UO 

«1 o 5 

r- 

r*. 



o 

ON 

vO 

VO 

fl) s 
oS 

co 


Ml* 

ON 

ON 

CO 


«H 






*—i 




^ C ~ 

<fl o . 









4-1 -rH O 

C 4-> t-l 

-3- 

H 

rH 

uo 

CO 

O’ 


«H 

O <fl 

vO 

, H 

rH 

o 

vn 

VO 


rH 

N 6 01 
•W k 41 

CM 

CM 

CM 

rH 

o 

CM 

CM 

CM 

44 O Ji 

0 4-10 

33 O C 

O 

O 

O 

o 

o 

O 

o 

O 

Q -H 









s-/ 









?H /~N 

nJ oi 



Mt 



tH 

CO 

CO 

O TJ "d 

OO 

00 

00 

oo 

00 

VO 

o 


•H « C 

r^* 

r- 

r^ 


r^ 

O0 

ON 

ON 

4-> o 3 

• 

• 

> 




44 ,-4 O 

ON 

ON 

On 

ON 

ON 

O 

CM 

co 

a> a. 

> w 

m 

CO 

co 

co 

co 


H 

*H 










C oi 
o a) 

•h a) 

4-1 44 

O 

o 

O 

O 

O 

o 

O 

o 

<fl 40 

4-1 <0 

O TS 

OS 


ON 


ON 


ON 


ON 


V) /—s 

CO CO 

Q) 0) 

ON 

ON 

vo 

VO 

CO 

CO 

00 

oo 

c x: 


r>* 

oo 

00 

U0 

UO 

00 

OO 

x o 
o C 

<r 

Ht 

mT 


<r 


<r 


•H «H 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

CM 

JZ ^ 

H 









VO 

VO 

UO 

o 

OO 

Mf 

vO 

CO 

uo 

UO 

ON 

C*\ 

• 

•- 


*. 

vo 

O 

CM 

UO 

VO 

co 

U0 

o 

rH 


i—1 

CO 


«H On «h| 

»—» VO ,—( 

CM O'") CM 

o o o 


UO 

o 

rH 

o 


00 

00 

CO 

CO 


CM 

O 

O 

ON 

OO 

ON 

ON 

co 

rH 

CM 

CM 

rH 

rH 

rH 

rH 


o o o o 

ON ON 


VO 

VO 

ON 

ON 

vo 

vO 

r- 

r-- 

<*• 




CM 

Csl 

CM 

CM 


a> 

U 

D W 
H 0) 

ctj a) 
u u 
0) bO 
CL Q) 

e x) 

d) N-' 

H 



0) 

AJ 

(0 T) 
bO C 
a> a) 

Vi rH 

bO 0Q 
bO 
< 


O 

CM 

i 

CO 




TABLE 14 (continued) 


•u 

C 

(/) 

fO 

o 

o 

NO 

CO 

4) 

3 

r-*- 

oo 

ON 

OO 

CM 

•>—< 

rH Q 

UO 

CM 

m 


CM 

t —< 

3 

- 

- 

- 

- 

- 

•H 

T5 ” 

in 

CO 


CM 


W 

o ®; 

ON 

o 

H 

00 

NO 

<D 

S 


m 

ON 

ON 

NO 

cd 




r— < 




CM 

CM 

CM 

OO 

r—♦ 

CM 

CM 

ON 



CO 

<r 

OO 

O 

*—l 

ON 

ON 

<f 

CM 

OO 

*— 4 

r~- 

H 

r—< 

NO 

O' 

CM 

NO 


CM 

CM 

NO 

r^. 

CO 

NO 

NO 

r—) 

»—1 

f—4 

r—) 

T - 4 

i — 1 


'-J 


CTJ 

^-4 

o 







•U 

•f—1 

O 






P 

U 

t—4 

O 

O 

cO 

m 

00 

o 

cd 


r-4 

r-< 

m 

o 

m 

N 

6 

V) 

CM 

Of 

o 

f-i 

T - 1 

•H 


O 






U 

o 

rC 

O 

o 

o 

o 

o 

O 

4-1 

o 






X 

<u 

p 







Q 

•H 







oo 

co 

CO 

O 

o 

CO 

o 

m 

NO 

NO 

r—t 

f— 1 1 

NO 

1 

T—1 

CM 

CM 

CM 

CM 

CM 

CM 

o 

O 

o 

o 

o 

o 

o 



94 








95 


TABLE 15 

SUMMARY OF RESILIENT MODULUS TEST RESULTS 


Percent 

Natural 

Sand 

Type of 

Sand 

Resilient 

Modulus 

77°F 

(psi) 

Resilient 

Modulus 

104°F 

(psi) 

0 

Crushed 

589,192 

190,354 

10 

Mason 

547,194 

164,722 


Concrete 

492,214 

99,412 

20 

Mason 

465,744 

199,522 


Concrete 

423,814 

126,833 

30 

Mason 

390,828 

147,414 


Concrete 

579,898 

140,431 















97 

Unconfined Creep-Rebound 

The unconfined creep-rebound test was conducted to evaluate the 
ability of the seven asphalt concrete mixtures to resist permanent 
deformation under severe loads. The creep-rebound test is one of the 
best laboratory procedures to determine rutting potential. The 
unconfined creep-rebound test was conducted at 77°F and 104°F and at 
loads that would produce a significant creep-rebound curve. The results 
of the unconfined creep-rebound test are presented in Table 16. Typical 
creep-rebound curves displaying axial deformation versus time are shown 
in the Appendix. 

A constant vertical load was desired to test all aggregate blends 
for each test temperature. The vertical load was selected to produce 
significant deformation in the stronger mixtures and not to overload the 
weaker mixtures. The initial vertical load was 40 psi for 77°F tests 
and 15 psi for 104°F tests. The 40 psi load worked satisfactorily until 
the 30 percent natural sand mixtures were tested. At the 30 percent 
natural sand content, the mixtures failed and the vertical load was 
decreased to 30 psi. A 15 psi vertical load was used to test the 0, 10, 
and 20 per-cent specimens at 104°F. This vertical load was decreased to 
10 psi foL 30 percent mason sand and 20 and 30 percent concrete sand 
mixtures because these asphalt concrete mixtures failed at the higher 
initial load. 

The results of the unconfined creep-rebound test were used to 
evaluate the seven asphalt concrete mixtures. The amount of axial 
deformation during the loading or creep phase indicated the ability of 
the mixture to resist deformation. Small axial deformations indicate 
stable mixtures with good resistance to deformation. The calculated 





creep modulus indicated the stiffness of the asphalt mixtures. High 
creep modulus values are desired to decrease rutting potential. The 
percent rebound or recovered deformation indicated the ability of the 
mixture to recover the traffic-induced deformation. High percent 
rebound values indicate that permanent deformation will be minimum. 

The amount of natural sand affected the test results of the 
unconfined creep-rebound test at both test temperatures. A relationship 
between the percentage of natural sand and the amount of axial 
deformation, creep modulus, and percent rebound was determined. The 
overall tendency was that the asphalt concrete mixtures weakened or 
increased in rutting potential as the natural sand content increased. A 
summary of the unconfined creep-rebound test values at 77°F and 104°F 
are presented in Tables 17-18. 

The creep-rebound values at 77°F indicated a significant relation¬ 
ship between the natural sand content and the creep - rebound properties. 
The axial deformation of the crushed limestone mixture (S-0) was 
0.0058 inches. The axial deformation for the mixtures containing 
natural sand increased as the percentage of natural sand increased. The 
average axial deformation was 0.0089 inches for 10 percent natural sand, 
0.0106 inches for 20 percent natural sand, and 0.0114 inches for 30 per¬ 
cent natural sand. The increase in axial deformation was 53.4 percent 
at 10 percent natural sand, 82.8 percent at 20 percent natural sand, and 
96.6 percent at 30 percent natural sand. Figure 46 displays the axial 
deformation values at 77°F versus the percent natural sand in mixture. 

The permanent deformation values also increased as the natural sand 
content increased. The permanent: deformation value for the S-0 blend 





99 


was 0.0039 inches. The average permanent deformation was 0.0069 inches, 
a 76.9 percent increase for 10 percent natural sand. The average 
permanent deformation was 0.0082 inches, a 110.3 percent increase for 
20 percent natural sand. The average permanent deformation was 
0.0092 inches, a 136.0 percent increase for 30 percent natural sand. 
Figure 47 displays the permanent deformation at 77°F versus the percent 
natural sand in mixture. 

The percent rebound values decreased as the natural sand increased. 
The percent rebound for the crushed limestone mixture (S-0) was 
33.2 percent. The average percent rebound was 27.5 percent for 10 per¬ 
cent natural sand, 23.3 percent for 20 percent natural sand, and 

20.4 percent for 30 percent natural sand. These values indicated that 
less deformation was recovered as the natural sand content increased. 

The creep modulus values decreased as the percentage of natural sand 
increased. The creep modulus values at 77°F are summarized in Tables 
17 and 19. The creep modulus value for the S-0 blend was 57,129 psi. 

The average creep modulus value was 36,899 psi for 10 percent natural 
sand, 31,085 psi for 20 natural sand and 22,553 psi for 30 percent 
natural sand. The decrease in creep modulus was significant as the 
natural sand content increased. The decrease in creep modulus was 

35.4 percent at 10 percent natural sand, 45.6 percent at 20 percent 
natural sand, and 60.5 percent at 30 percent natural sand. Figure 48 
displays the creep modulus values versus the percent sand in mixture. 

The creep-rebound values at 104°F also indicated a significant 
relationship between the natural sand content and the creep-rebound 
properties. The test results are not as consistent as the values at 



100 


77°F, but do show the expected tendencies. Since different vertical 
loads were used, a direct comparison of deformations cannot be 
graphically analyzed. The tendencies observed in the 77°F tests were 
also evident in the axial and permanent deformation values. In both 
creep-rebound properties, the deformation increased as the natural sand 
content increased. 

The creep modulus values also decreased as the percentage of natural 
sand increased. The creep modulus values at 104°F are summarized in 
Tables 18-19. The creep modulus value for the S-0 blend was 23,872 psi. 
The average creep modulus was 15,816 psi for 10 percent natural sand, 
12,549 psi for 20 percent natural sand, and 10,216 psi for 30 percent 
natural sand. The decrease in creep modulus values was 33.8 percent at 
10 percent natural sand, 47.5 percent for 20 percent natural sand, and 
57.2 percent at 30 percent natural sand. The decrease in creep modulus 
at 104°F is also significant. Figure 49 displays the creep modulus 
values versus the percent natural sand in mixture. 






UNCONFINED CREEP-REBOUND TEST RESULTS 


101 


•O U 


§ 

o 

c 

0) 

o 

o 


O' 

lO 

on 

00 

iH 

m 


m 

CO 

O 

oo 

o 

ON 

H 

cM 

VO 

00 

CO 

vO 

ON 

-O 

Vi 

CO 

CO 

CO 

CO 

CO 


CO 

Csi 

CO 

CSI 

CSI 


Q) <U 
03 P. 


C 

P O 

C "H 

0) P 

C « 

S H 

S Vi 
Vi O 


Oi 0) 

a 


(/) 

ON 

o 

00 

CM 

CM 

CM 

CO 

o 



co 

a> 


CO 

CO 

cO 

co 

CO 

OO 


m 

<t 

m 

a 

o 

o 

o 

o 

o 

o 

O 

o 

o 

o 

o 

o 

c 

•h 

o 

o 

o 

o 

o 

o 

O 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 


M 

P. 3 

/~s 

r^ 

ON 

CM 

in 

« r-l 


o 

O 

r«* 

CO 

0) 3 

to 

CM 

CM 

O' 

O 

Vi T) 

a 

- 

- 

- 

■» 

U O 

v_^ 

VO 

vO 

00 

CO 

X 



VO 

m 

CM 


c 



o 

o 

ON 

in 

ro 

f—1 

P U] 

r-~ 


n 

in 


«o a) 

O 

o 

o 

O 


6 -C 

O 

o 

O 

o 

X 

vi o 

• 

• 

• 

• 

< 

o c 

o 

o 

o 

o 


4-1 -H 
<u ' 

a 


ON 

co 

VO 

VO 

VO 

CO 

ON 

in 

CM 

-4- 

m 

ON 

r—l 

CM 

m 

o 

vO 

On 

in 

m 

CO 

vO 

CM 

00 

CM 

m 

oo 

in 

CM 

CM 

CO 

CO 

rM 

»—i 

T—4 


r- 

in 

CO 

OO 



n 


in 

oo 

ON 


vo 

r- 

O 

o 

o 

o 

O 

o 

o 

o 

o 

o 

o 

o 

o 

o 


o o o o o o o 


R) v—- 

O -O tJ 
"H RJ 01 
WOP. 

<D 

> 


O 


oomininoommin 

<}■ <f f-H r—< rH <t ^ 1 —* »—* 


0) 

/—\ 

u 


3 

1 

4J 

to 

id 

0) 

u 

a) 

0) 

V4 

a 

bO 

a 

d) 

0) 

*o 

H 

V-/ 


r-* r- 


o 


<* 

o 


<* 

o 


r- 


r» 

r- 


o 


o 


o 


<v 

u 

cd TJ 
to c 
v d> 

H r-l 

b0 CQ 
b0 
< 


O 

i 

00 









TABLE 16 (continued) 


102 


T3 

P 

C 

r- 

oo 


o 

o 

vO 

vo 

CM 

i—11 

m 

P 

<1) 




i • i 

1 • 



i • 


• 

o 

o 

»—i 

ON 

in 

i CO 1 

♦ VO 

r-» 

vO 

* 


oo 

JO 

U 

CM 

1—1 

co 

T—1 

1—1 

CM 

i-H 

CM 

CM 

co 

a) 

CD 

P, 












c 

« o 
e -h 


a> 

4J W 

ON 

H 

r—4 

vo 

vo 

oo 

m 

CO 

i~j 

CO 

C 

c0 <D 

r- 

o 

in 

co 


vO 

o 

in 

m 


c0 

£ rC 

o 

r—4 

o 

1 o * 

• r— • 

O 

H 

» o 

o 

o 

e 

^ o 

o 

o 

o 

• o • 

« o 

o 

o 

» o 

o 

o 


O P 




• 





• 

• 

<D 

4-4 -H 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 


PM (1) ^ 
Q 


p- 

in 

p 

ON 

VO 

r- 

VO 

<D 

r—i *rl 

o 

OO 

*—( 

1 ON 

<D 

P in 

CO 

VO 

vO 

i r- 

U 

X> (X 

*» 

- 


- 

o 

o ^ 

t—4 

uo 


o 



CO 

CM 

1—1 

T —i 


r—i 

ON 

CM 

vo 

CM 

vO 

CO 

CM 

r—i 

1 UO 

o 

OO 

i-H 

r- 

VO 

• f—i 


vo 


m 

ON 

i-H 

H 

o 

rH 

CM 

1— i 

1—1 

i—i 

H 



P 






/N 

/^s 









o 




id 


«d 

cd 




cd 






H 

vo 

ON 


UO 


vx 



VO 


O 

o 

o 

r—i 

u <n 

O 

CM 



o 



r- 

O 

CM 


r-* 

r- 

r-** 

03 

id a> 

i —i 

r~i 

o 

1 

f—f 

• 

1 

i—i 

O 

i—1 

i 

o 

o 

o 

•t-4 

£ rC 

o 

O 

o 

t 

o 

• 

1 

o 

o 

O 

i 

o 

o 

o 

X 

u o 



• 


. 



. 




• 



< 

O P 

4-1 -H 

o 

o 

o 


o 



o 

o 

o 


o 

o 

o 


0) -w 

o 


CO /-“N 


o 

*i-4 

T) -H 

cd w 

o 

o 

UO 

UO 

UO 

uo 

o 

o 

O 

O 

uo 

O 

o 

o 

4-> 

o cx 



t—1 

H 

1—1 

r—i 


CO 

co 

CO 

1—1 

r—1 

1—1 

r —1 

u 

0) 

> 

H-) ^ 
















0) ^ 

U U-i 

P • 
u co 

CO 0> 

u a) 
a) u 
P- bO 
6 a> 
a; *0 
H *— 


r*". 

r-* 



a> 

■u 

CO T3 
b£) c 
a) a> 

U i—4 
&0 CQ 
60 
< 


CM 

i 

CO 


s: 

ro 

i 

co 


(a) Sample failure-overloaded 





TABLE 16 (continued) 


103 


T> 

% 

u 

c 

a) 

c«"» 

OO 


CM 

OO 


00 

o 

VO 

i 

m 

OO 

O 

o 



m 

CO 

o 

iHJ 

CO 

CM 

» o 



£> 

a) 

d) 

CM 

r-1 

co 

CM 

CO 

CM 

CM 

CO 

CO 

in 

>3- 

& 

cx 













c 

9) 

C 


c 

o 


O CO 

. <0 o> 

« e x. 

e n o 

HOC 
9) 4-1 

Hi HI v 

0 


CO 

CO 

r —1 

vD 

r-> 

ON 

VO 

m 




m 

00 

m 

vO 

<f 


00 

m 

r- 



O 

o 

o 

o 

o 

o 

o 

o 

• o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

■ o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 


cx 

w 

3 

X"-\ 

CM 

H 

OO 


-o- 

CO 


CM 

m 


r- 

a) 

H 

*rH 

CM 


m 

vO 

co 

CO 

rH 

ON 

O' 

CO 


a) 

3 

10 

CM 

hs. 

o\ 



T — ( 

m 

VO 

» OO 

VO 


u 


cx 

• 


- 



•> 



1 


o 

O 

V-/ 


*—t 


CO 

VO 

CM 

OO 

Ox 

o 

VO 






CO 

i—( 

i — < 

»—< 

CO 

CM 

CO 


»—< 

i — < 


c 



o 

•rl 

/^l 

CO 

»—t 

ON 

vO 

00 

rH 

x> 

w 


o 

r- 

CO 

to 


tO 

a> 

O 

«-l 

o 

O 

O 

•t-l 

£ 

rC 

o 

o 

o 

O 

O 


u 

o 



• 



< 

o 

c 

o 

o 

o 

O 

O 


<XI 

9) 

Q 

•rl 







CO 


rM 

CO 

t—t 


r—4 

r~- 

OO 

o 

T-1 

00 


H 


m 

T—i 

*—1 

o 

l 

»—1 

O 

o 

O 

o 

o 

< 

O 

O 

o 

o 

o 

o 


O 

o 

o 


cd 

O X) >H 
•W cfl (/) 
X> O CX 
H iJ —' 
0) 

> 


O o m 

-tt tH 


inmooomooo 

*—t «—t -Cf r—t c—t r—i t—l 


<U ^~\ 
H Cx 
3 ■ 

CJ U) 

to a) 
U <U 
<U 5-t 
CX bt> 
6 <U 
0) T) 
H 



0) 

V 

<0 X3 

bO C 

9) 9) 

U r-t 

b£) 0 
bO 
< 


O 

i 

CO 


o 

CM 

i 

CO 


(a) Sample failure-overloaded 







TABLE 16 (continued) 


104 


T) 

£ 

3 

O 

rO 

<1> 

CtZ 


4J 

c 

a) 

o 

u 

Q) 

P- 


m 

in 

vo 

in 

vo 

m 

CM 

CM 

m 

in 

CSI 

CM 

CO 

CO 

CO 


c 


■U 

C 

<D 

O 

•H /—v 

U « 

T — l 


vO 

CM 

CO 


C 

<d <i> 

OO 

vO 

OO 

vO 

in 


<d 

E .jC 

1 o 

o 

o 

O 

o 

o 

a 

u o 

» O 

o 

o 

o 

o 

o 

u 

o c 


• 


• 

• 

• 

0) 

4-1 -H 

O 

o 

o 

o 

o 

o 

Oh 

dj ^ 

Q 








01 

a. 3 ^ 

ON 


o 

O 



a> .— i 

O 

CO 

o 

CO 



a) 3 oi 

i 


r- 

vO 


vo 

U T5 (X 

• 

- 

- 

- 

- 

- 

o o ^ 

m 

O0 

r—1 

00 

00 

o 

X 

CM 

CM 

CM 



»—i 



o 

•r—1 


<d 

VO 

1—1 

CM 

o 

m 

r — < 

4J 

w 

av 

00 

T—1 

o> 

ON 

r- 

<d 

<d 

a) 

« o 

o 

1—1 

o 

o 

o 

•r—1 

e 

rC 

• o 

o 

o 

o 

o 

o 

X 

u 

a 


• 

• 

• 

* 


< 

o 

c 

o 

o 

o 

o 

o 

o 


4-1 

a> 

Q 

•H 

V-' 








<d 

o 

•rl 

T3 T-l 

<d w 

O 

O 

o 

o 

o 

O 

■u 

o a- 

Mf 

CO 

CO 

CO 

f—1 

r—1 

u 

Q) 

> 

»-4 








(1) /-N 

P tu 


3 • 

4J W 

03 a> 
p a> 
a) w 

CX b0 

6 a; 
a) x> 
H "~" 


r-- 

r- 




O 


o 


o 


<1) 

U 

03 X) 
b0 C 
<D <D 
M i—l 
bD CQ 
b£> 

< 


U 

co 

i 

CO 


*o 

0) 

X) 

<d 

o 

f-H 

u 

0) 

> 

o 

I 

a) 

p 

3 


03 

<4-1 


o- 

E 

<d 

cn 




105 


w 

-4 

I 



"O +j 












S c 

3 4) 

CM 

CO 

VO 

uo 

OO 

OO 

1 ^ 

H 

VO 

1 ^ 


O O 

CO 

CM 

CM 

r- 

o 

m 

CO 

o 

Q 

o 


J3 H 

CO 

CO 

CM 

CM 

CM 

CM 

1 CM 

CM 

CM 

1 CM 


4) 4) 

04 a. 
























C 












U> O 

C *fl 

4) -U W 

cr\ 

o 

00 

ON 

o 

CO 

CM 

VO 


CM 


C <4 4> 

CO 

r>* 

VO 

vO 

ON 

fN 

00 

o 


ON 


g B X 

O 

o 

o 

o 

o 

oj 

o 


o 

o 


a n o 

O 

o 

o 

O 

o 

°l 

o 

o 

o 

o 


(j O c 

• 

• 


• 

• 


. 





4) 4-1 <r-l 

Pu 4) w 

o 

o 

o 

o 

o 

ol 

o 

o 

ol 

o 


Q 











o 
























r*^ 












H 












< 












H 

<fl 

P. 3 ^ 











CO 

w 

ON 

T—1 


<JN 

CO 

v£> 

m 


» — i 

co 

f-* 

4) i— 1 *»-l 

CM 

o 

ON 

on 

CM 


00 

CM 

oo 

in 


4) 3 (0 

r—1 

00 

ON 

OO 



o 

oo 

CM 

m 

Q 

N U U. 

» 

• 

• 

- 

« 


- 



55 

O O ^ 

rv 

in 


vO 

oo 

CO 

rH 

ON 

m 

CM 

p 

s 

m 

CO 

CO 

CO 

CM 

CO 

CO 

f—< 

CM 

CM 

O 












CQ 












W 












Pi 












(L 












u 












w 

c 











O 

o *-* 

10 

CO 

»—i 


ON 


00 

vO 

*—1 


<t 

Q 

f-4 *-> 4) 

m 

on 

oo 

CO 

H 

ON 

o 

CO 

ON 

»—1 

w 

B) «4 £, 

O 

o 

O 

o 

f-4 

o 

f—< 


o 

f—1 

55 
►—< 

•h a o 
* ^ c 

O 

o 

O 

o 

o 

o 

o 

o 

o 

o 

flH 

< O -H 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

2i 












O 

O 

4) 

Q 























P 












&h 












o 













H 

cd 




x> 






X) 

«2 

O T3 >H 




£ 



c 



c 


tI U W 

o 

o 

o 

cd 

o 

o 

cd 

o 

o 

cd 

8 

POP. 


<*■ 


CO 



CO 

CO 

CO 

CO 

CO 

H i-J 

0) 




<#> 



<#> 



dP 

> 




o 



o 



o 






»"*4 



CM 



CO 






u 



u 



u 






o 



o 



o 









<4-1 



<4-1 


4-1 



0) 



<U 



0) 



X) 


u 

0) 


u 

<D 


4-> 

0) 


O 13 

<D 


0) 

00 


Q) 

tO 


0) 

oo 


c 


c 

u 

cd 

c 

u 

cd 

c 

u 

cd 


4) <4 

W 

o 

o 

n 

o 

o 

u 

o 

o 

u 


(X C/i 


w 

c 

<0 

w 

c 

0) 

w 

c 

d) 


►> 

V4 

<d 

o 

> 

Cd 

o 

> 

cd 

o 

> 


H 

o 

£ 

CJ 

< 

s: 

u 

< 

£ 

o 

< 


4J .-1 












C fl 3 

4) H C 

o 

o 



o 



o 




O 3 <4 
pun 
« <4 

O. Z 


t—< 



CM 



CO 







SUMMARY OF UNCONFINED CREEP-REBOUND TEST AT 104°F 


106 


X) V> 

e c 

04 

oo 

00 


o 

CO 

vO 

3 a) 


• 


• 

. 



o o 

oo 

O' 

On 

VO 

m 

On 

<T 

rO VI 

CO 

CM 

CM 

CM 


CM 

CO 

<V CL) 

05 CL 

N—C 

e 

v> o 

C -H /-V 
(1 L It 

04 

o 

m 

On 

CO 

ON 

vO 

C «fl 0) 

CO 

in 

m 

VO 



m 

« e .c 

O 

o 

o 

O 

o 

o 

o 

e vi o 

• o 

o 

o 

O 

o 

o 

o 

C o c 


• 

• 

• 

• 

• 

. 

0) 4-1 •<-( 

o 

o 

o 

O 

o 

o 

o 

CV <U 

a 

w 

a 3 ^ 

CM 

ON 

CM 

r^ 

1—1 

*—1 

»—4 

H) 1 —1 -r-l 

o- 

o- 

m 

O 

ON 

oo 

m 

a> 3 w 

co 


H 

o- 

CO 

o 

CO 

Vi X) CL 


- 

•» 

- 

- 

- 

- 

U O 

CO 

vO 

m 

CM 

CM 

1 — 1 

ON 

s 

04 

H 


»— 1 

i-H 

«— 1 


c 

o 

"H W 

CM 

CM 


CM 

CM 

o 

m 

r-l 4J d) 

in 

O- 


ON 

O- 


oo 

ai fl 

o 

o 

O 

O 

O 

o 

o 

■h e o 

o 

o 

O 

O 

o 

o 

o 

X Vi C 


• 

• 

• 


• 


< O *H 

o 

o 

o 

o 

o 

o 

o 

4-1 w 

<u 

Q 

.-1 

a) 

o x> -.-i 
•h a! w 

m 

in 

in 

in 

o 

o 

o 

pop, 

t—4 

«—4 

r—4 

r-l 

t—4 

*— 1 

i—i 

Vi >4 -w' 

0) 

> 

4-1 

O X! 

c 

T> 

0) 

X 

r 

0) 

■u 

0) 

Vi 

c 

0) 

4-> 

0) 

Vi 

c 

0) 

•u 

0) 

Vi 

a) a3 

w 

o 

O 

O 

o 

o 

O 

CV CO 

3 

w 

C 

w 

C 

M 

C 

>1 

Vi 

cd 

O 

<d 

o 

<d 

o 

H 

O 

asj 

O 


o 


o 

Percent 

Natural 

Sand 

O 

10 


20 


30 












TABLE 19 


108 


SUMMARY OF CREEP MODULUS VALUES 


Percent Creep Creep 


Natural 

Sand 

Type of 
Sand 

Modulus 

(@77°F) 

Decrease 

(percent) 

Modulus 

(@104°F) 

Decrease 

(percent) 

0 

Crushed 

57,129 


23.872 


10 

Mason 

35,801 

37.3 

16,479 

31.0 


Concrete 

37.997 

33.5 

15.152 

36.5 


Average 

36,899 

35.4 

15,816 

33.8 

20 

Mason 

28,723 

49.7 

12,707 

46.8 


Concrete 

33.446 

41.5 

12.391 

48.1 


Average 

31,085 

45.6 

12,549 

47.5 

30 

Mason 

19,824 

65.3 

11,081 

53.6 


Concrete 

25.281 

55.7 

_JLJJ>1 

60.8 


Average 

22,553 

60.5 

10,216 

57.2 






0 10 20 

Percent Sand in Mixture 

Figure 48. Creep Modulus Values at 77°F 


Creep Modulus psi (Thousands) 



d 10 20 

Percent Sand in Mixture 

Figure 49. Creep Modulus Values at 104°F 








CHAPTER VI 


SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 
Summary 

This laboratory study was conducted to evaluate the effects of 
natural sands on the engineering properties of asphalt concrete 
mixtures. This research program consisted of a review of available 
literature and existing data, and a two-phase laboratory study on 
laboratory - produced specimens. Conventional and state-of-the-art 
testing procedures and equipment were used to determine the effects of 
natural sands on asphalt concrete mixtures. The objective of this 
research was to examine the engineering properties of the asphalt 
concrete mixtures and to set quantitative limits of natural sand to 
prevent unstable mixtures and reduce rutting potential. 

The review of the literature and existing data indicated that the 
quality and size of the aggregate had a tremendous effect on the 
properties of asphalt concrete mixtures. Several laboratory research 
studies had been conducted comparing natural or uncrushed aggregates to 
crushed coarse and tine aggregates. The conclusions of these laboratory 
studies indicated that stability and strength properties of mixtures 
decreased as the percentage of uncrushed aggregates increased. These 
studies also indicated that replacing natural sand materials with 
crushed sands would increase the resistance to permanent deformation in 


110 





Ill 


asphalt concrete pavements. 

The first phase of this laboratory study evaluated the physical 
properties of the materials used in this study. Aggregate gradations 
were computed to produce aggregate blends that were as consistent as 
possible. Asphalt concrete mix designs were conducted on the seven 
aggregate blends to select optimum asphalt contents. 

The aggregate blends were produced using 0, 10, 20, and 30 percent 
natural sand. These blends were fabricated as close as possible to the 
target gradation. However, the aggregate blends for the 20 and 30 per¬ 
cent natural sand contents did have some variation, especially at the 
No. 30 sieve. A definite hump developed at the No. 30 sieve when these 
gradations were plotted on standard semi-log graphs and graphs with 
sieve sizes raised to the 0.45 power. This hump in the gradation curves 
indicated that asphalt mixtures with 20 and 30 percent natural sand 
contents were sensitive and tender. 

The optimum asphalt content was determined for each aggregate blend 
using the Marshall mix design procedure. Several trends were evident 
from the mixture properties at the optimum asphalt contents. The 
optimum asphalt content decreased as the percentage of natural sand 
increased. The stability values were also effected by the percentage of 
natural sand; the stability values decreased as the percentage of 
natural sand increased. Another relationship that was observed was a 
decrease in voids in mineral aggregate as the percentage of natural sand 
increased. Each of these trends or relationships indicated that the 
quality and durability of the asphalt concrete mixture both decreased as 
the percentage of natural sand increased. 





112 


The second phase of the laboratory study evaluated the effects of 
natural sands on asphalt concrete mixtures using state-of-the-art 
testing equipment. Specimens were produced for each aggregate blend at 
the optimum asphalt content and evaluated with the Marshall procedure, 
indirect tensile test, resilient modulus test, and unconfined creep- 
rebound test. 

The Marshall mix properties were determined so these values could be 
analyzed with the more modern test procedures. The test properties 
determined in Phase II agreed with the trends and relationships observed 
in Phase I. The mix properties including stability, voids filled with 
asphalt, and voids in mineral aggregate decreased as the percentage of 
natural sand increased. These Marshall properties indicated that 
natural sand materials lowered the strength properties and would affect 
the durability of the asphalt mixture by decreasing the asphalt content 
and void properties. 

The indirect tensile test was conducted to determine the tensile 
strength properties of the seven asphalt concrete mixtures. The tensile 
strength values were effected by the percentage of natural sand and the 
test temperature. The relationship was evident that the amount of 
natural sand controlled the strength properties of the mixtures. As the 
natural sand content increased, the tensile strength decreased. The 
test temperature significantly affected the tensile strength; at 104°F 
the tensile strength was three times less than the tensile strength 
values at 77°F. The tensile strength was much lower at 104°F, which 
indicated rutting potential would be greater at higher pavement 


temperatures. 





113 


The resilient modulus test was conducted to determine the relative 
quality of the asphalt concrete mixtures. The resilient modulus values 
produced in this study were very inconsistent. The ASTM procedure used 
to determined the resilient modulus relies heavily on measuring very 
small deformations. This measurement is very sensitive and produces 
large variations in the results. The consistency of the resilient 
modulus values determined in this study was not satisfactory. The 
unreliability of resilient modulus values has also been documented by 
Brown and Foo (7). 

The unconfined creep-rebound test is considered one of the best 
laboratory procedures to determine rutting potential in asphalt concrete 
mixtures. This test procedure evaluated the ability of the mixtures to 
resist permanent deformation under severe loads. The unconfined creep- 
rebound values indicated that the rutting potential of asphalt concrete 
mixtures increased as the percentage of natural sand increased. The 
axial and permanent deformations were larger at higher natural sand 
contents. The creep modulus value decreased as the percentage of 
natural sand increased. The stiffness of the mixtures was much lower at 
104°F, which indicated rutting potential was greater at higher pavement 
temperatures. 

Conclusions 

Based on the results of the laboratory investigation which included 
the literature review and two-phase laboratory study, the following 
conclusions were made on the effects of natural sands on engineering 
properties of asphalt concrete mixtures: 





114 


1. The use of natural sand materials decreased the stability and 
strength characteristics of asphalt concrete mixtures. 

2. Replacing natural sand materials with crushed sand materials 
increased the resistance to permanent deformation in asphalt 
concrete mixtures. 

3. High natural sand contents, 20 percent and higher, caused aggregate 
blending problems. These natural sand contents produced gradations 
with high percentages of material passing the No. 30 sieve. 

4. Aggregate gradations with 20 and 30 percent natural sand produced a 
definite hump at the No. 30 sieve when using a grading curve with 
the sieve sized raised to the 0.45 power. 

5. Optimum asphalt content values decreased as the percentage of 
natural sand increased. The asphalt content required to produce a 
mixture at 4 percent voids total mix was much lower for a mixture 
with high natural sand content. Lower asphalt contents produce a 
less durable pavement. 

6. Marshall stability values decreased as the percentage of natural 
sand increased. The stability values were significantly reduced at 
the 20 and 30 percent natural sand contents. The stability values 
decreased to a level that was below the minimum 1800 lbs requirement 
at 30 percent natural sand. 

7. The voids in mineral aggregate (VMA) decreased as the percentage of 
natural sand increased. 

8. The indirect tensile results indicated a reduction in mixture 
strength as the percentage of natural sand increased. The 
temperature of the indirect tensile test significantly effected the 






115 

tensile strength value. The higher temperature produced lower 
strength values. This test procedure indicated a definite trend 
when evaluating the natural sand content. 

9. The resilient modulus test results were very inconsistent and 
indicated no trend. This test procedure was not a good test 
procedure to evaluate the effects of natural sands in asphalt 
concrete mixtures. The variation in test results for duplicate 
samples was very large. Deformation of the specimens may have 
occurred during the first test which caused the variation in the 
second resilient modulus value. 

10. The unconfined creep-rebound test results indicated a strong 
relationship between the percentage of natural sand and rutting 
potential. The axial and permanent deformation values increased 
tremendously as the natural sand content increased. The creep 
modulus value decreased significantly as the percentage of natural 
sand increased. The creep-rebound test values were significantly 
affected at the 20 and 30 percent natural sand contents. 

11. All laboratory test results indicated that asphalt concrete mixtures 
with all crushed aggregates had higher strength properties and would 
resist potential rutting better than mixtures containing natural 
sand materials. Asphalt concrete mixtures containing more than 

20 percent natural sand appeared to have tremendous potential to 


deform under severe loads. 




116 


Recommendations 

Based on the conclusions derived from the results of this laboratory 

study, the following recommendations were made: 

1. To maximize the reduction in rutting potential for heavy duty 
pavements, all crushed aggregate should be used in the asphalt 
concrete mixture. • 

2. The maximum allowable limit for the natural sand content for heavy 
duty pavements should be less than 20 percent by weight. A 
conservative b’ t practical maximum limit should be 15 percent 
natural sand. 

3. Unconfined creep-rebound and indirect tensile tests should be used 
in conjunction with the Marshall procedure to analyze asphalt 
concrete mixtures in order to fully evaluate the engineering 
properties. 

4. Aggregate gradations should be plotted on a gradation curve with the 
sieve sizes raised to the 0.45 power to evaluate the tenderness of 
the mixture. 

5. Further laboratory studies should be conducted to evaluate the 
effects of other characteristics of natural sand materials in 
asphalt concrete mixtures. Aggregate type, angularity, particle 
shape, and gradation of the natural sand should be analyzed in more 
detail. 

6. Field investigations should be conducted to verify field performance 
with laboratory data. 




117 


REFERENCES 


1. Al-Hussaini, M. M., and Townsend, F C. (1973, May). "Tensile 
Testing of Soils -- A Literature Review, Miscellaneous Paper S-73-24, 
U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS. 

2. American Association of State Highway and Transportation Officials. 
(1988, November). "Report of AASHTO Joint Task Force on Rutting," 
Washington, DC 

3. American Society of Testing and Materials. 1990. 1990 Annual Book 

of ASTM Standards . Vol 0403, Philadelphia, PA. 

4. Anderton, G. L. (1990, January). "Evaluation of Asphalt Binder 
Modifiers," Miscellaneous Paper GL-90-1, U.S. Army Engineer Waterways 
Experiment Station, Vicksburg, MS. 

5. Brown, E. R. (1974, December). "Evaluation of Asphalt Concrete 
Pavements Subjected to T-38 Aircraft Loadings," Thesis, Mississippi 
State University, Mississippi State, MS. 

6. Brown, E. R. (1987, February). "Mix Design and Construction of 
Asphalt Concrete to Support High Tire Pressures," AASHTO/FHWA Symposium 
on High Pressure Truck Tires, Austin, TX. 

7. Brown, E. R. and Foo, K. Y. (1989, October). "Evaluation of 
Variability in Resilient Modulus Test Results" (ASTM D 4123), ASTM 
Journal of Testing and Evaluation, Philadelphia, PA. 

8. Button, J. W., and Perdomo, D. (1990, January). "Influence of 
Aggregate on Rutting in Asphalt Concrete Pavements," Transportation 
Research Board, Washington, DC. 

9. Department of the Army. (1984, September). "Bituminous 
Intermediate and Wearing Courses for Airfields, Heliports and Heavy-Duty 
Pavements (Central-Plant Hot-Mix), Corps of Engineers Guide 
Specification, Military Construction CEGS-02556, Washington, DC. 

10. Department of the Army and the Air Force. (1987, July). Technical 
Manual TM 5-822-8/AFM 88-6, Bituminous Pavements Standard Practice . 
Washington, DC. 

11. Department of Commerce. (1962, May). "Aggregate Gradation for 
Highways," Washington, DC. 




118 


12. Department of Defense, MIL-STD-620A. (1966, January). Military 
Standard Test Methods of Bituminous Paving Materials . Washington, DC. 

13. Development of Gyratory Testing Machine and Procedures for Testing 
Bituminous Paving Mixtures . (1962, February). Technical Report 3-595, 
U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS. 

14. Federal Highway Administration. (1988, March). "Asphalt Concrete 
Mix Design and Field Control," FHWA Technical Advisory T 5040.27, 
Washington, DC. 

15. Foster, C. R. 1986. "The Effects of Voids in Mineral Aggregate on 
Pavement Performance," National Asphalt Pavement Association, 

Information Series 96/86, Riverdale, MD. 

16. Grau, R. W. 1979. "Utilization of Marginal Construction Materials 
for LOC," Technical Report GL-79-21, U.S. Army Engineer Waterways 
Experiment Station, Vicksburg, MS. 

17. Griffith, J. M., and Kallas, B. F. 1957. "Aggregate Voids 
Characteristics in Asphalt Paving Mixes," Highway Research Board 
Proceedings, Vol 36. 

18. Griffith, J. M., and Kallas, B. F. 1958. "Influence of Fine 
Aggregates on Asphalt Concrete Paving Mixtures," Highway Research Board 
Proceedings, Vol 37. 

19. Hanson, D. I. (1984, May). "Asphalt Mix Control in New Mexico," 
Proceedings FHWA Tri-Regional Pavement Rehabilitation Conference, 
Oklahoma City, OK. 

20. Herrin, M. and Goetz, W. H. 1954. "Effect of Aggregate Shape on 
Stability of Bituminous Mixes," Highway Research Board Proceedings. 

21. Investigation of the Design and control of Asphalt Paving Mixtures . 
(1948, May). Technical Memorandum No. 3-254, U.S. Army Engineer 
Waterways Experiment Station, Vicksburg, MS. 

22. Kalcheff, I. V., and Tunnicliff, D. G. 1982. "Effects of Crushed 
Stone Aggregate Size and Shape on Properties of Asphalt Concrete," 
Proceedings AAPT, Vol 51. 

23. Marker, V. (1977, June). "Tender Mixes: The Causes and 
Prevention," The Asphalt Institute, Information Series No. 168. 

24. Marks, '. J., Monroe, R. W., and Adam, J. F. (1990, January). 

"The Effects of Crushed Particles in Asphalt Mixtures," Transportation 
Research Board, Washington, DC. 

25. McRae, J. L. 1965. "Gyratory Testing Machine Technical Manual," 
Vicksburg, MS. 




119 


26. McRae, J. L. and Foster, C. R. (1959, April). "Theory and 
Application of a Gyratory Testing Machine for Hot-Mix Bituminous 
Pavement," Miscellaneous Paper 4-333, U.S. Army Engineer Waterways 
Experiment Station, Vicksburg, MS. 

27. Page, G. C. (1984, May). "Rutting-Causes and Preventions," 
Proceedings FHWA Tri-Regional Pavement Rehabilitation Conference, 
Oklahoma City, OK. 

28. Pavement Mix Design Study for Very Heavy Gear Loads Pilot Test 
Section . (1962, February). Technical Report 3-594, U.S. Army Engineer 
Waterways Experiment Station, Vicksburg, MS. 

29. Shklarsky, E., and Livneh, M. 1964. "The Use of Gravels for 
Bituminous Mixtures," Proceedings AAPT, Vol 33. 

30. Wedding, P. A., and Gaynor, R. D. 1961. "The Effects of Using 
Gravel as the Coarse and Fine Aggregate in Dense Graded Bituminous 
Mixtures," Proceedings AAPT, Vol 30. 

31. _ (1982, April). "Rutting Investigation," 

Materials Laboratory, Wyoming State Highway Department, Cheyenne, WY. 

32. _ 1984. "Asphalt Pavement Rutting-Western 

States," Western Association of State Highway and Transportation 
Officials, WASHTO Executive Committee. 






APPENDIX A: UNCONFINED CREEP-REBOUND CURVES 


120 



















































































































































































155 



O 

CM 


O 

CJ> 


O 

ID 


O 

CO 


QO)vO»-E«)+' , "OC 


C or 0 CO 


Time, Minutes 

Figure 84. Creep-Rebound Curve for S-2C6 Sample at 104°F